UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Consumer automobile choice and gasoline consumption Kanetkar, Vinay 1984

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1984_A4_6 K35.pdf [ 6.26MB ]
Metadata
JSON: 831-1.0096024.json
JSON-LD: 831-1.0096024-ld.json
RDF/XML (Pretty): 831-1.0096024-rdf.xml
RDF/JSON: 831-1.0096024-rdf.json
Turtle: 831-1.0096024-turtle.txt
N-Triples: 831-1.0096024-rdf-ntriples.txt
Original Record: 831-1.0096024-source.json
Full Text
831-1.0096024-fulltext.txt
Citation
831-1.0096024.ris

Full Text

CONSUMER AUTOMOBILE CHOICE AND GASOLINE CONSUMPTION by VI NAY KANETKAR B. Arch., Indian I n s t i t u t e of Technology, 1974. M. Arch., The U n i v e r s i t y of B r i t i s h Columbia, 1979 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE (BUS. ADMIN.) in THE FACULTY OF GRADUATE STUDIES ( F a c u l t y of Commerce and Bus. Admin.) We accept t h i s t h e s i s as conforming to the r e q u i r e d standard. THE UNIVERSITY OF BRITISH COLUMBIA J u l y , 1984 © Vinay Kanetkar, 1984 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y a v a i l a b l e for reference and study. I further agree that permission for extensive copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head of my department or by h i s or her representatives. I t i s understood that copying or pub l i c a t i o n of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. BepartmottJt of- ^ Cotnnmsir&L The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date ' Kayjk Ofr**-i i ABSTRACT The i n c r e a s i n g emphasis on g a s o l i n e c o n s e r v a t i o n programs i n d i c a t e s the importance of understanding consumer g a s o l i n e consumption p a t t e r n s . Consumers may reduce g a s o l i n e consumption by reducing automobile weight, payload, d i s t a n c e t r a v e l l e d , and f r i c t i o n a l l o s s e s . I t i s argued that understanding consumer d e c i s i o n s about the s i z e o f household automobile f l e e t and car weight w i l l improve the c u r r e n t understanding of g a s o l i n e consumption and p o s s i b l y l e a d to c o n s e r v a t i o n programs which w i l l reduce g a s o l i n e demand. A review of econometric l i t e r a t u r e at the consumer l e v e l i n d i c a t e d that income, f a m i l y composition, p r i c e of g a s o l i n e , c e n t r a l c i t y r e s i d e n c y , and employment s t a t u s are important determinants of the number of c a r s owned by consumers. A review of marketing s t u d i e s i n d i c a t e d that a t t i t u d e s and p e r s o n a l i t y f a c t o r s p r o v i d e a d d i t i o n a l understanding regarding consumer automobile c h o i c e s . F i n a l l y , a conceptual model was used to i n t e g r a t e these approaches, suggesting that the choice of automobiles i s determined by s i t u a t i o n a l , market r e l a t e d f a c t o r s , consumer needs as w e l l as consumer c h a r a c t e r i s t i c s . The conceptual model i m p l i e s s e v e r a l e m p i r i c a l hypotheses. For example, i t was p o s t u l a t e d that income, f a m i l y s i z e , and employment s t a t u s w i l l have a p o s i t i v e impact on the number of  ca r s owned by a f a m i l y , while s i z e of urban centr e and g a s o l i n e p r i c e s w i l l have a negative impact. I t was a l s o expected that income and f a m i l y s i z e would p o s i t i v e l y a f f e c t automobile s i z e . The age, s i z e of urban c e n t r e and energy consciousness of the head of the household w i l l n e g a t i v e l y a f f e c t t h i s v a r i a b l e . L o g i t models c o n s t r u c t e d to p r e d i c t the number of c a r s owned by a Canadian f a m i l y gave s i g n i f i c a n t r e s u l t s f o r the above f a c t o r s . In a d d i t i o n , the l i f e - c y c l e stage and the proxy v a r i a b l e s f o r g a s o l i n e p r i c e s had n o n - l i n e a r e f f e c t s on t h i s v a r i a b l e . Models p r e d i c t i n g the average car weight, however, suggested t h a t a t t i t u d e s towards c o n s e r v a t i o n , and p a r t i c i p a t i o n i n g a s o l i n e c o n s e r v a t i o n e f f o r t s were a l s o strong p r e d i c t o r s . Thus, t h i s r e s e a r c h concluded that demographic and s i t u a t i o n a l f a c t o r s are l i n k e d to the number of ca r s owned, while t r a v e l needs and a t t i t u d i n a l f a c t o r s are l i n k e d to car weight. P r o f . John D. Claxton, J u l y , 1984. TABLE OF CONTENTS ABSTRACT i i TABLE OF CONTENTS i i i LIST OF TABLES i v LIST OF FIGURES i v ACKNOWLEDGMENTS v CHAPTER I Consumer G a s o l i n e Conservation 1 1.1 I n t r o d u c t i o n to Gas o l i n e Conservation 2 1.2 B a s i c s of G a s o l i n e Consumption 4 1.3 Consumer G a s o l i n e Conservation A c t i o n s 9 1 .4 Conclusi o n s 16 CHAPTER II T h e o r e t i c a l Development of Automobile Ownership 18 2.1 I n t r o d u c t i o n to Car Ownership Research 19 2.2 T h e o r e t i c a l C o n s i d e r a t i o n s 21 2.3 Measurement and E s t i m a t i o n of Car Demand Fun c t i o n .... 25 2.4 Econometric Examples of Automobile Demand Functions .. 32 2.5 Marketing Examples of Automobile Ownership S t u d i e s ... 47 2.6 An I n t e g r a t e d Model of Automobile Ownership 55 2.7 P o s s i b l e Research Areas 62 CHAPTER III Data Sources, Hypothesis and A n a l y t i c a l Methods 65 3.1 I n t r o d u c t i o n 66 3.2 The ECCP Sample to the Po p u l a t i o n Estimate Comparison 67 3.3 M o d e l l i n g Number of Cars Owned 74 3.4 M o d e l l i n g Car S i z e Preference 80 3.5 Methods of A n a l y s i s 82 CHAPTER IV Result and Conclusion 88 4.1 I n t r o d u c t i o n 89 4.2 Models R e l a t i n g to the Number of Cars owned 90 4.3 Models P r e d i c t i n g Average Car. Weight 104 4.4 Comprehensive Model of Consumer Car F l e e t Choice .... 113 4.5 P o l i c y I m p l i c a t i o n s and Future Research 116 REFERENCES 119 APPENDIX I A U t i l i t y Maximization Model 124 APPENDIX II E s t i m a t i o n Procedure f o r L o g i t Model 128 LIST OF TABLES Table 1.1 MPG and Auto Weight R e l a t i o n s h i p 5 Table 1.2 R e l a t i o n s h i p between Auto Weight and Auto S i z e 6 Table 1.3 The G a s o l i n e Conservation O b j e c t i v e s and Outcomes . 11 Table 1.4 Conservation A c t i v i t i e s and Expected Savings 12 Table 2.1 Consumer D e c i s i o n s and Choice of Dependent V a r i a b l e s 26 Table 2.2 Summary of S t u d i e s P r e d i c t i n g Number of Cars Owned , 35 Table 2.3 Summary of S t u d i e s P r e d i c t i n g Automobile Purchase Choice 40 Table 2.4 Automobile and Consumer C h a r a c t e r i s t i c s I n t e r - r e l a t i o n s 61 Table 3.1 Average Car Ownership per Car Owning Household .... 69 Table 3.2 Sample New Car Purchases and P o p u l a t i o n New Car S a l e s 70 Table 3.3 F l e e t E f f i c i e n c y Comparison 71 Table 3.4 Ownership P a t t e r n Comparison 74 Table 3.5 A t t i t u d e and P a r t i c i p a t i o n Items from ECCP Data Base 82 Table 3.6 D e s c r i p t i o n of Independent V a r i a b l e s 83 Table 4.1 MLE f o r P r e d i c t i n g P r o b a b i l i t i e s of Number of Cars Owned, Demographic and S i t u a t i o n a l V a r i a b l e s . 92 Table 4.2 Impact of L i f e - c y c l e on Number of Cars Owned 96 Table 4.3 Impact of G a s o l i n e P r i c e s on Number of Cars Owned . 98 Table 4.4 MLE f o r P r e d i c t i n g P r o b a b i l i t i e s of Number of Cars Owned, S e l e c t e d Set of V a r i a b l e s 100 Table 4.5 T r i a l Sample Models f o r Average Car Weight 106 Table 4.6 Model P r e d i c t i n g Average Car Weight 107 LIST OF FIGURES F i g u r e 2.1 An I n t e g r a t e d Model of Automobile Ownership 55 F i g u r e 4.1 F a c t o r s I n f l u e n c i n g Automobile F l e e t Choice 114 V ACKNOWLEDGMENTS I wish to express my s i n c e r e g r a t i t u d e to my t h e s i s a d v i s o r s , P r o f . John D. Claxton and Pro f . Doyle L. Weiss. T h e i r experience and patience made t h i s p o s s i b l e . I a l s o wish to express my g r a t i t u d e to P r o f . W i l l i a m Waters, f o r s e r v i n g on my t h e s i s committee and p r o v i d i n g i n s i g h t f u l comments. I f e e l indebted to P r o f . I l a n V e r t i n s k y , P r o f . Gordon McDougall, P r o f . Donald Wehrung, Prof. Kenneth MacCrimmon, Pro f . G e r a l d Gorn, Pr o f . W i l l i a m Stanbury, P r o f . Peter Nemetz, and Pr o f . C h a r l e s Weinburg. Through t h e i r work, I gained experience with a very v a l u a b l e range of a n a l y t i c a l methods, h o p e f u l l y evident in t h i s document. My f i n a l note of g r a t i t u d e goes to a l l my f r i e n d s f o r p r o v i d i n g both a s t i m u l a t i n g and p e a c e f u l s o c i a l environment i n Vancouver, and to my f a m i l y , e s p e c i a l l y my b r o t h e r s V i j a y and V r a j a l a l and my mother, f o r t h e i r encourgement and mental support. Vancouver, J u l y , 1984. 1 CHAPTER I Consumer G a s o l i n e C o n s e r v a t i o n . 1.1 I n t r o d u c t i o n to Ga s o l i n e Conservation 1.2 B a s i c s of G a s o l i n e Consumption 1.3 Consumer G a s o l i n e Conservation A c t i o n s 1.4 Conclus i o n s 2 1.1 I n t r o d u c t i o n to G a s o l i n e Conservation In the l a s t decade car p o o l i n g , tune-up d r i v e s , and c a r l e s s - d a y s have emerged as new approaches to reduce g a s o l i n e consumption in the the p r i v a t e t r a n s p o r t a t i o n s e c t o r . The b a s i c o b j e c t i v e s behind these emerging approaches have been to improve f l e e t e f f i c i e n c y and/or to reduce f l e e t usage. T h i s research focuses on the f a c t o r s i n f l u e n c i n g the e f f i c i e n c y of the p r i v a t e s e c t o r automobile f l e e t . From the consumer resea r c h p e r s p e c t i v e , f l e e t e f f i c i e n c y can be viewed in terms of (1) the f u e l e f f i c i e n c y of c a r s a v a i l a b l e to consumers, (2) consumer automobile ownership and purchase d e c i s i o n s , and (3) consumer maintenance p r a c t i c e s employed to maintain or upgrade mechanical e f f i c i e n c y . The purpose of t h i s r e s e a r c h i s to understand the f a c t o r s i n f l u e n c i n g consumer ownership d e c i s i o n s . T h i s t h e s i s i s organized i n t o four c h a p t e r s . In chapter I, the impact of automobile weight, payload, machine e f f i c i e n c y and t r a v e l d i s t a n c e on g a s o l i n e consumption are examined and summarized. I t . i s suggested that g a s o l i n e consumption can be minimized by pursuing v a r i o u s a c t i v i t i e s that minimize f l e e t weight, payload and t r a v e l d i s t a n c e , and f r i c t i o n a l l o s s e s . The i m p l i c a t i o n s of 30 consumer a c t i o n s are summarized and organized around these four o b j e c t i v e s . T h i s chapter concludes that the a c t i o n s r e q u i r e d to maximize the machine e f f i c i e n c y of i n d i v i d u a l l y owned ca r s are w e l l documented, although consumer 3 purchase d e c i s i o n s r e g a r d i n g f l e e t e f f i c i e n c y are not w e l l understood. In chapter I I , the t h e o r e t i c a l b a s i s u n d e r l y i n g the f a c t o r s a f f e c t i n g f l e e t ownership i s d e r i v e d from the consumption f u n c t i o n theory, and from past e m p i r i c a l r e s e a r c h i n the economic and marketing l i t e r a t u r e . In a d d i t i o n , a model presented i n t h i s chapter suggests that consumers' f l e e t e f f i c i e n c y d e c i s i o n s are determined by s i t u a t i o n a l or market r e l a t e d f a c t o r s , consumers' f l e e t usage needs, and consumer c h a r a c t e r i s t i c s . A n a t i o n a l sample c o l l e c t e d by McDougall e t . al.(l979) c o n t a i n s e x t e n s i v e i n f o r m a t i o n about households, t h e i r automobile f l e e t s , and f a m i l y c h a r a c t e r i s t i c s . T h i s data provided the e m p i r i c a l component of the present r e s e a r c h . Chapter III examines the McDougall sample i n terms of the number of c a r s owned and automobile f l e e t e f f i c i e n c y , as w e l l as compares i t to the Canadian p o p u l a t i o n . As a r e s u l t of these comparisons, i t i s concluded that the sample can be used to c o n s t r u c t and t e s t models concerned with f l e e t e f f i c i e n c y . In chapter IV, the sample data are used to c o n s t r u c t models which p r e d i c t (1) fam i l y ownership (consumer c h o i c e ) of one, two and three or more automobiles, and (2) mean car weight. I t appears from the s t a t i s t i c a l r e s u l t s that the number of automobiles owned i s mostly determined by the consumers' demographic c h a r a c t e r i s t i c s . However, mean car weight or machine e f f i c i e n c y depends upon the consumers' energy 4 c o n s e r v a t i o n p a r t i c i p a t i o n and a t t i t u d e s . 1.2 B a s i c s of G a s o l i n e Consumption In the simplest e n g i n e e r i n g terms, an automobile' i s a machine that i s used to p r o p e l a cargo and i t s own weight a c e r t a i n d i s t a n c e . The g a s o l i n e consumed in order to accomplish t h i s p r o p u l s i o n should be i n v e r s e l y r e l a t e d to v e h i c l e weight, cargo and / or "payload" weight and machine e f f i c i e n c y ; i t w i l l be d i r e c t l y p r o p o r t i o n a l to the t o t a l d i s t a n c e t r a v e l l e d . Each of these v a r i a b l e s are reviewed in t h i s s e c t i o n . Impact of Weight. I t i s not s u r p r i s i n g to f i n d that l a b o r a t o r y evidence shows that v a r i a t i o n i n the r a t e of g a s o l i n e consumption (miles per g a l l o n or l i t r e s per 100km) i s a f u n c t i o n of automobile weight. For example, using data from Consumer Reports between 1972 and 1980, i t appears that g a s o l i n e consumption (MPG) i s l i n e a r l y r e l a t e d to automobile weight with R 2's ranging from 0.62 to 0.88 (Table 1.1). I t i s a l s o c l e a r that weight, engine s i z e , as w e l l as the p h y s i c a l volume of the v e h i c l e are h i g h l y r e l a t e d (Table 1.2). The r e s u l t s i n Table 1.1 and 1.2 imply that v e h i c l e g a s o l i n e consumption can be understood by f o c u s s i n g on the f a c t o r s which i n f l u e n c e an automobile's weight and s i z e . Impact of Payload. The r e l a t i o n s h i p between e f f i c i e n c y and automobile weight i s simple and l i k e l y holds i n most f i e l d d r i v i n g s i t u a t i o n s . However, v a r i a n c e i n t h i s r e l a t i o n s h i p Table 1 . 1 MPG and Auto Weight 1 R e l a t i o n s h i p Model year Sample s i z e M i l e s per g a l l o n on c i t y roads M i l e s per g a l l o n on highways Const. C o e f f . of weight R 2 Const. C o e f f . of weight R 2 1 972 36 23.66 -0.0036 0.882 39.32 -0.0053 0.790 1 974 45 24.32 -0.0036 0.732 41 .87 -0.0057 0.820 1 976 57 25.54 -0.0039 0.686 43. 1 4 -0.0062 0.728 1 978 52 30.42 -0.0055 0.773 51 .93 -0.0085 0.747 1 980 64 35.33 -0.0069 0.620 63.67 -0.0120 0.720 Source: Consumer Reports, 1972-1980 'Weight i s measured i n pounds Q> Table 1.2 R e l a t i o n s h i p between Auto Weight 1 and Auto S i z e ( t - s t a t i s t i c s are pro v i d e d i n parentheses) Model year Sample s i z e Constant C o e f f . f o r auto s i z e 2 C o e f f . f o r engine s i z e 3 R 2 1 972 36 1323.3 0.34/10 3 (2.12) 6.67 (9.41) 0.928 1974 45 456.4 0.00141 (9.32) 2.82 (4.33) 0.973 1976 57 922. 1 0.00107 (6.76) 3.18 (4.29) 0.937 1 978 52 532. 1 0.00147 (9.25) 2.25 (3.36) 0.960 1980 64 713.1 0.00125 (7.70) 2.66 (3.75) 0.899 1Weight measured i n terms of pounds 2Auto s i z e measured i n terms of c u b i c inches 3Engine s i z e measured i n terms of c u b i c inches 7 a r i s e s from the u n c e r t a i n t i e s a s s o c i a t e d with payload s i z e . For example, commuting to work may mean c a r r y i n g fewer passengers s i n c e i t i s u n l i k e l y that a l l family members are employed at the same work p l a c e . On the other hand, i f the car i s used at work, the consumer may be c a r r y i n g a d d i t i o n a l work r e l a t e d cargo. F u r t h e r , the s o c i a l s i g n i f i c a n c e of f a m i l y shopping i s l i k e l y to i n f l u e n c e both the number of passengers and the amount of cargo c a r r i e d d u r i n g shopping t r i p s . In a d d i t i o n , d i f f e r e n t r e l a t i o n s h i p s w i l l e x i s t f o r other usages such as r e c r e a t i o n a l , s o c i a l a c t i v i t i e s and personal business. To summarize these c o m p l e x i t i e s , f a m i l y s i z e may provide a u s e f u l measure; with l a r g e r f a m i l i e s expected to c a r r y l a r g e r payloads. Impact of Machine E f f i c i e n c y . In a l a b o r a t o r y s e t t i n g the r e l a t i o n s h i p between g a s o l i n e consumption and (1) a i r and r o l l i n g r e s i s t a n c e s , and (2) braking and t r a n s m i s s i o n l o s s e s are s t a b l e and n e g a t i v e . However, the i n t e n s i t y of these r e l a t i o n s h i p s depend upon the u n p r e d i c t a b l e c o n d i t i o n s of wind, r a i n , road s u r f a c e and the l i k e . T h i s i m p l i e s that the mechanical e f f i c i e n c y w i l l not only d i f f e r from place to p l a c e and d r i v e r to d r i v e r , but a l s o during d i f f e r e n t seasons of the year. As i m p l i e d , some of these f r i c t i o n a l l o s s e s can be i n f l u e n c e d by v a r i o u s consumer or d r i v e r a c t i o n s . For example, a i r r e s i s t a n c e can be reduced by reducing automobile speed. S i m i l a r l y , r o l l i n g r e s i s t a n c e i s i n f l u e n c e d by the type of t i r e s used, as w e l l as the t i r e pressure l e v e l . Braking and 8 t r a n s m i s s i o n l o s s e s can be reduced by r e g u l a r tune-ups and l u b r i c a t i o n and by lowering a c c e l e r a t i o n and v e l o c i t y . To conclude, although c l i m a t i c f a c t o r s a f f e c t machine e f f i c i e n c y , consumer a c t i o n s such as tune-ups, t i r e p r e s s u r e awareness, equipment s e l e c t i o n and d r i v i n g " s t y l e " can r e s u l t i n g a s o l i n e s a v i n g s . Sources of T o t a l D i s t a n c e . Another f a c t o r which has an obvious impact on g a s o l i n e consumption i s the t o t a l d i s t a n c e t r a v e l l e d . However, the f a c t o r s i n f l u e n c i n g v e h i c l e usage, and hence, t r a v e l l i n g d i s t a n c e are numerous and d i v e r s e . For example, i t can be argued that as p o p u l a t i o n d e n s i t y i n c r e a s e s , d i s t a n c e s between v a r i o u s p o i n t s of i n t e r e s t (and hence d i s t a n c e t r a v e l l e d ) may decrease. F u r t h e r , as p o p u l a t i o n d e n s i t y i n c r e a s e s , the a v a i l a b i l i t y of p u b l i c t r a n s p o r t a t i o n w i l l i n c r e a s e and t h i s may a f f e c t the frequency of automobile usage. The i m p l i c a t i o n i s t h a t , i n g e n e r a l , high d e n s i t y p o p u l a t i o n areas would r e q u i r e l e s s d r i v i n g than suburban or r u r a l p o p u l a t i o n areas. I t i s l i k e l y that the v a r i e t y of car uses (commuting, l e i s u r e , v a c a t i o n i n t e r e s t s , etc.) as w e l l as the number of f a m i l y members of d r i v i n g age w i l l a l s o a f f e c t the t o t a l d i s t a n c e t r a v e l l e d . To summarize, automobile g a s o l i n e consumption depends upon automobile weight, payload, machine e f f i c i e n c y and t o t a l d i s t a n c e t r a v e l l e d . Most of these f a c t o r s are d i r e c t l y a f f e c t e d by consumer a c t i o n s . For example, v e h i c l e weight i s determined by consumer choice s i n c e weight i s c l o s e l y r e l a t e d to car s i z e 9 and s i z e i s an important c o n s i d e r a t i o n i n car purchase d e c i s i o n s . On the other hand, consumer s i t u a t i o n s that determine payload and d i s t a n c e t r a v e l l e d are not l i k e l y to be c e n t e r e d on g a s o l i n e consumption c o n s i d e r a t i o n s , at l e a s t in the short term. I n f l u e n c i n g machine e f f i c i e n c y by means of c a r e f u l d r i v i n g and maintenance p r a c t i c e s however, i s l i k e l y to be motivated by a d e s i r e to minimize the c o s t of g a s o l i n e consumpt i o n . 1.3 Consumer G a s o l i n e Conservat ion Act ions The preceeding d i s c u s s i o n i n d i c a t e s that consumers can reduce g a s o l i n e consumption by pursuing four b a s i c a c t i o n s . F i r s t , consumers can reduce the weight of an automobile by purchasing a smaller c a r . Second, consumers can minimize the payload f o r each t r i p . T h i r d , consumers can reduce the number of t r i p s and t o t a l mileage by p l a n n i n g multipurpose t r i p s or by using p u b l i c t r a n s p o r t a t i o n . And f o u r t h , consumers can improve machine e f f i c i e n c y by c a r e f u l maintenance and d r i v i n g p r a c t i c e s . Each of these o b j e c t i v e s are commented on i n t h i s s e c t i o n . The outcomes of the v a r i o u s consumer c o n s e r v a t i o n a c t i v i t i e s d i s c u s s e d above can be c l a s s i f i e d i n terms of (1) a gain i n energy e f f i c i e n c y , (2) a c u r t a i l m e n t i n energy consumption or (3) a demand s h i f t , i . e . , s w i t c h i n g from one energy form to another (McDougall e t . a l . , 1981; Stern and Gardner, 1981). These outcomes o f t e n c o n f l i c t with other 10 d e s i r a b l e outcomes (Mufti and Munson, 1979). For i n s t a n c e , outcomes of consumer comfort, s a f e t y , f l e x i b i l i t y and t r a v e l time o f t e n c o n f l i c t with a c h i e v i n g the best mechanical e f f i c i e n c y ( W a l l i n and Wright, 1974). I l l u s t r a t i v e examples of c o n s e r v a t i o n a c t i v i t i e s and energy r e l a t e d outcomes as w e l l as other outcomes are p r o v i d e d i n Table 1.3. For example, to minimize auto weight, a consumer can purchase a s m a l l e r v e h i c l e . Although the a c t i o n w i l l improve e f f i c i e n c y , i t w i l l a l s o decrease passenger s a f e t y . S i m i l a r l y , multipurpose t r i p s w i l l reduce t o t a l t r a v e l , but at the c o s t of f l e x i b i l i t y . A recent O n t a r i o M i n i s t r y of T r a n s p o r t a t i o n study l i s t s some 20-25 a c t i v i t i e s which reduce g a s o l i n e consumption. A m o d i f i e d l i s t of these a c t i v i t i e s c a t e g o r i z e d by the four g e n e r a l o b j e c t i v e s p r o v i d e s some i n t e r e s t i n g o b s e r v a t i o n s (Table 1.4). F i r s t , of 28 a c t i v i t i e s summarized, 19 deal with e n g i n e e r i n g estimates of energy savings through a v a r i e t y of mechanisms designed to maximize the machine e f f i c i e n c y of c u r r e n t l y owned automobiles. Second, estimates of savings from human f a c t o r s such as d r i v i n g h a b i t s , multi-purpose t r i p s or automobile purchase d e c i s i o n s are not d i s c u s s e d . T h i r d , means f o r minimizing automobile weight are not d i s c u s s e d . Expected savings from such a c t i v i t i e s can be i n the range of 20%. Fourth, and f i n a l l y , the savings i n d i c a t e d i n Table 1.4 are of a t e c h n i c a l nature and do not o f f e r a measure of consumer a c t i o n s or usage behaviour. Table 1.3 The G a s o l i n e C o n s e r v a t i o n O b j e c t i v e s and Outcomes Object i v e s Examples Energy outcomes Other outcomes M i n i m i z e f i x e d weight Purchase s m a l l e r automobile Increase i n e f f i c i e n c y Decrease i n s a f e t y M i n i m i z e p a y l o a d C a r r y fewer t h i n g i n automobile I n c r e a s e i n e f f i c i ency Decrease i n f l e x i b i l i t y M i n i m i z e t o t a l m i l e s d r i v e n Use m u l t i -purpose t r i p s C u r t a i l m e n t and a demand s h i f t Decrease i n f l e x i b i l i t y M i n i m i z e f r i c t i o n a l l o s s e s Use appro-p r i a t e t i r e s Some g a i n i n ef f i c i e n c y Need t o change t i r e s as weather changes T a b l e C o n s e r v a t i o n A c t i v i t i e s (from O n t a r i o Energy 1 .4 and Expected S a v i n g s D r i v e - S a v e R e p o r t ) 1 2 Consumer a c t i v i t y E x p e c t e d f u e l s a v i n g s B a s i c s a v i n g mechanism O b j e c t i v e : M i n i m i z e Weight 1. Purchase s m a l l e r c a r than p r e s e n t one 1% l e s s f o r ever y 451bs weight reduc-t i o n or 20% f o r c h anging from 1980 compact to subcompact S m a l l e r v e h i c l e s are g e n e r a l l y l i g h t e r and hence w i l l consume on an average 1% l e s s g a s o l i n e f o r every 451bs c u t i n w e i g h t . I f v e h i c l e i s newer t h e n , the engine may have s l i g h t l y h i g h e r ef f i c i e n c y . 2. Purchase newer but same s i z e model No d a t a ava i l a b l e See above 3. Add s m a l l e r v e h i c l e t o c a r f l e e t No d a t a I f m o t i v a t i o n b e h i n d a d d i n g v e h i c l e i s to m i n i m i z e g a s o l i n e c o n s u m p t i o n , ( e . g . , u s i n g s m a l l auto when pa s s e n g e r s a r e one or two w h i l e u s i n g b i g c a r f o r l o n g j o u r n e y s ) then t h e r e might be some g a s o l i n e s a v i n g s O b j e c t i v e : M i n i m i z e Cargo 4. M i n i m i z e 1% f o r every T h i s i s a t a c t i c a l a c t i v i t y t o reduce t h i n g s 451bs weight consumption. Even i f one assumed t h a t and reduced most a u t o s c a r r y an e x t r a weight of people 451bs, e x p e c t e d s a v i n g i s l a r g e . However, t h e r e i s no base case t o compare i t t o . O b j e c t i v e : M i n i m i z e T o t a l D i s t a n c e T r a v e l l e d Reduce t r a v e l by managing t r a v e l L e v e l of c u r t a i l m e n t w i l l be r e f l e c t e d i n s a v i n g s A l t h o u g h t r a v e l p l a n n i n g can save 1 t o 5% i n g a s o l i n e , t h e r e i s l i t t l e i n f o r m a t i o n about p r o c e s s and consumer r e a c t i o n s . S a v i n g o c c u r s from two s o u r c e s : f i r s t l e s s d r i v i n g and second fewer t i m e s c a r i s s t a r t e d . 13 Table 1.4 (cont i nued) Use a l t e r n a -t i ve means of t r a v e l , i f p o s s i b l e Same as above There are some i n d i c a t i o n s that consumers are p a r t i c i p a t i n g in t h i s form of c o n s e r v a t i o n , e.g. g r e a t e r use of p u b l i c t r a n s i t , and g r e a t e r b i c y c l e s a l e s . However, the demand s h i f t might have occured through p r o h i b i t i v e c o s t of auto ownership. Hence, there i s need for a c a r e f u l e v a l u a t i o n of the cause and e f f e c t s . Ride shar ing, vanpool-ing and var i e t y a of other forms Claims of savings range from 50 to 60% Although these are a l t e r n a t i v e means of t r a n s p o r t a t i o n , they are more f l e x i b l e than p u b l i c t r a n s i t . Success of these types of a c t i v i t i e s depends, however, upon marketing e f f o r t s s p e c i f i c a l l y p o s i t i o n i n g c a r p o o l i n g as a f l e x i b l e mode of t r a n s p o r t a t i o n Adjust work schedule so that stop-goes are redu-ced 2 tO 6% d u r i n g peak hours Automobile i s d r i v e n more e f f i c i e n t l y while there i s l e s s peak demand on highway systems. However, t h i s type of a c t i v i t y r e q u i r e s c o o p e r a t i o n between v a r i o u s o r g a n i z a t i o n w i t h i n a community and some t r i p s may become more time consuming. Move c l o s e r work place to No data ava i l a b l e Consumers i n d i r e c t l y c u r t a i l t h e i r d r i v i n g . However, they a l s o i n t r o d u c e demand f o r housing c l o s e r to t h e i r work p l a c e , i . e . u s u a l l y w i t h i n a b u s i n e s s d i s t r i c t . O b j e c t i v e : Maximize Machine Ef f i c iency 10. Forgo A/C or minimize i t s use 1% f o r weight reduct ion 3 to 4% for use Urban d r i v i n g with a i r - c o n d i t i o n e r i n c r e a s e s consumption by 8 to 10%. Penalty i n c r e a s e s s h a r p l y f o r o l d e r and heavier c a r s . 11. Buy manual st e e r ing over power s t e e r i n g 1 to 3% f o r s m a l l e r c a r s T h i s o p t i o n i s only u s e f u l f o r s m a l l e r car s. 12. Choose manual t ransmi s-s i o n over automat i c T y p i c a l l y 8% f o r s m a l l e r c a r s Automatic t r a n s m i s s i o n i s f u e l e f f i c i e n t and u s e f u l f o r l a r g e r c a r s but not f o r smaller c a r s . I 4 T a b l e 1.4 ( c o n t i n u e d ) 13. Turn o f f engine when stopped f o r more than 10 sec and not i n t r a f f i c No d a t a ava i l a b l e 10 s e c . of i d l i n g consumes more f u e l than to r e s t a r t . 14. Use moderate and c o n s t a n t a c c e l e r -a t i o n r a t e s No d a t a ava i l a b l e More power output means g r e a t e r f u e l c onsumption. 0-50 km/hour i n 15 s e c . appears t o s e r v e as a good " r u l e of thumb". 15. C r u i s e a t 60 to 70 km/hour 15 t o 20% at 80km/h, 30% or more at 40km/h H i g h e r and lower speeds a r e g e n e r a l l y l e s s f u e l e f f i c i e n t 16. A l l o w speed t o f a l l o f f on h i l l s 1 t o 2% C o m b i n a t i o n of p o t e n t i a l and auto energy i s used f o r c r u i s i n g . Not u s e f u l on highways. 17. Use SAE 5W30 or 10W30 o i l a l l year round 3 t o 4% i n c o l d e r Low v i s c o s i t y h e l p s d u r i n g c o o l e r engine o p e r a t i o n . S h o r t urban type t r i p s can have the g r e a t e s t b e n e f i t . Not u s e f u l i f the s u r r o u n d i n g t e m p e r a t u r e i s above 16 C (60 F) or w i t h o l d e r a u t o s . 18. Use f r i c t i o n modi f i e d or s y n t h e t i c o i l s 2 t o 3% o v e r a l l s a v i n g O l d e r v e h i c l e s may not have any p o s i t i v e e f f e c t . 19. Use r a d i a l or P-metr i c r a d i a l t i r e s 4 to 5% and a d d i t i o n a l 1 t o 1.5% f o r P-met r i c B e t t e r d r y and wet t r a c t i o n . Optimum r e s i s t a n c e and l o n g e r t r e a d l i f e p r o v i d e r e a s o n a b l e s a v i n g s . O l d e r c a r s may not be d e s i g n e d f o r such t y p e s of t i r e s . 20. Keep t i r e p r e s s u r e at opt imum 1% f o r e v e r y 14 kpa(2 p s i ) T i r e p r e s s u r e at optimum l e v e l reduces r e s i s t a n c e , improves a c c e l e r -a t i o n , as w e l l as c o r n e r i n g and b r a k i n g t r a c t i o n . I t a l s o reduces t r e a d wear. 15 T a b l e 1.4 (cont i nued) 21. Remove snow t i r e s e a r l y i n s p r i n g 4 t o 5% approx i m a t e l y Snow t i r e s have h i g h e r r e s i s t a n c e on d r y e r r o a d s . 22. Engine tune-ups r e g u l a r l y 4 t o 5% f o r m o d e r a t e l y out of tune a u t o s Once a year appears t o t o be optimum f r e q u e n c y f o r tune-ups. 23. Use b l o c k h e a t e r i n w i n t e r 4 t o 8% dur i ng warm-up Engine s t a r t a b i l i t y i s improved, i n t e r i o r heat i s a v a i l a b l e q u i c k l y and o v e r a l l f u e l economy i s improved. However, t h e r e might be some demand s h i f t i n terms of g r e a t e r use of e l e c t r i c i t y . 24. Use f u e l s u ggested by manufact-ure No e f f e c t on consumpt i o n The h i g h e r octane i n premium f u e l , t o g e t h e r w i t h h i g h e r v o l a t i l i t y , i n c r e a s e s r e s p o n s i v e n e s s , improves s t a r t a b i l i t y and reduces engine run-on. However, t h i s does not improve f u e l economy i n a s h o r t e x p e r i m e n t a l s e t t i n g . I t i s c l a i m e d t h a t u n leaded f u e l l e d a u t o s r e q u i r e l e s s m aintenance. Thus, i n the l o n g run t h i s may h e l p i n f u e l s a v i n g . 25. Purchase r e t r o f i t f u e l sav i ng d e v i c e s No proven e f f e c t 26. Keep h i ghway speed moderate 10% p e n a l t y f o r each 10 . 10 km/hour i n c r e a s e i n speed above 50-60 'km/hour see a c t i v i t y 15 27. Keep s i d e w i ndows c l o s e d 1% a t 65 km/h 2% a t 105 km/h I n c r e a s e i n a i r - r e s i s t a n c e , i n c r e a s e s f u e l c onsumption. 28. Remove' roof r a c k s , m i r r o r s when not i n use 5% t o 30% p e n a l t y when i n use Same as a c t i v i t y 26. but l o s s e s are g r e a t e r on highways. 1 6 To summarize, consumers can pursue four major o b j e c t i v e s to reduce g a s o l i n e consumption, minimize weight, payload and t o t a l t r a v e l d i s t a n c e , and maximize machine e f f i c i e n c y . Although gains from e f f i c i e n c y r e l a t e d a c t i v i t i e s are w e l l documented, t h e i r b e n e f i t s on c o n s e r v a t i o n are small when compared to consumer a c t i v i t i e s a s s o c i a t e d with the r e d u c t i o n of automobile weight or s i z e . As a r e s u l t , a major rese a r c h concern, when attempting to understand g a s o l i n e c o n s e r v a t i o n determinants, i s the issue of v e h i c l e weight and how ownership d e c i s i o n s i n f l u e n c e t h i s v a r i a b l e . T h i s i s the focus of t h i s t h e s i s . 1.4 Con c l u s i o n s The i n c r e a s i n g emphasis p l a c e d on g a s o l i n e c o n s e r v a t i o n programs i n d i c a t e s the importance of understanding consumer g a s o l i n e consumption p a t t e r n s . S t a r t i n g with an e n g i n e e r i n g model, i t i s argued that MPG i s n e g a t i v e l y r e l a t e d to automobile weight and machine e f f i c i e n c y . F u r t h e r , g a s o l i n e consumption i s a f u n c t i o n of MPG and t o t a l d i s t a n c e t r a v e l l e d . Although the g a s o l i n e consumption f u n c t i o n can be c o n s t r u c t e d a c c u r a t e l y i n l a b o r a t o r y s e t t i n g s , the model would be l e s s u s e f u l in p r e d i c t i n g f i e l d consumption because of unknown b e h a v i o u r a l and environmental f a c t o r s . By m i n i m i z i n g four b a s i c o b j e c t i v e s - automobile weight, payload, d i s t a n c e t r a v e l l e d , and f r i c t i o n a l l o s s e s - consumers might reduce g a s o l i n e consumption. These o b j e c t i v e s were used 17 to s t r u c t u r e and d i s c u s s a v a r i e t y of s p e c i f i c c o n s e r v a t i o n a c t i o n s . While review of these o b j e c t i v e s suggested that the f a c t o r s a f f e c t i n g machine e f f i c i e n c y are w e l l understood, i s s u e s r e l a t e d to usage and ownership are not w e l l understood. One b e h a v i o u r a l a c t i v i t y t h a t p r o v i d e s c o n s i d e r a b l e p o t e n t i a l f o r g a s o l i n e savings i s consumers' d e c i s i o n s about f l e e t e f f i c i e n c y . More s p e c i f i c a l l y , understanding consumer d e c i s i o n s about s i z e of automobile f l e e t and f l e e t weight c o u l d improve the understanding of programs aimed at reducing g a s o l i n e consumption through b e h a v i o u r a l means. These areas w i l l be explored i n the next chapter. 18 CHAPTER II T h e o r e t i c a l Development of Automobile Ownership 2.1 I n t r o d u c t i o n to Car Ownership Research 2.2 T h e o r e t i c a l C o n s i d e r a t i o n s 2.3 Measurement and E s t i m a t i o n of Demand Fun c t i o n 2.4 Econometric Examples of Demand Function 2.5 Marketing Examples of Automobile Ownership S t u d i e s 2.6 An I n t e g r a t e d Model of Automobile Ownership 2.7 P o s s i b l e Research Areas 19 2.1 I n t r o d u c t i o n to Car Ownership Research A new need to study automobile ownership has emerged from recent i n t e r e s t i n energy c o n s e r v a t i o n . Economists, t r a n s p o r t a t i o n planners and marketers, however, had been studying t h i s t o p i c for more than 30 years. The purpose of t h i s chapter i s to review much of t h i s past r e s e a r c h and then to provide a t h e o r e t i c a l b a s i s f o r i d e n t i f y i n g the f a c t o r s a f f e c t i n g the demand f o r automobiles. T h i s i s accomplished by (1) a review and a n a l y s i s of the consumption f u n c t i o n theory, (2) a review of consumer l e v e l econometric s t u d i e s of automobile ownership, and (3) a review of the r e l a t e d marketing l i t e r a t u r e . In s e c t i o n 2.2 the consumption f u n c t i o n theory i s reviewed and d i s c u s s e d . The theory not only p r o v i d e s i n s i g h t s about f a c t o r s a f f e c t i n g ownership but i t a l s o suggests s e v e r a l p o s s i b l e measures of car ownership. In s e c t i o n 2.3 a d i s c u s s i o n of these measures of ownership suggests that ownership must be measured by the number of c a r s owned as w e l l as car c h a r a c t e r i s t i c s . F i n a l l y , to estimate a demand equation f o r the number of c a r s owned, two p o s s i b l e techniques are examined; a d i s c r e t e c h o i c e model (using l o g i t a n a l y s i s ) , and d i s c r i m i n a n t a n a l y s i s . In s e c t i o n 2.4 the review of f i v e econometric s t u d i e s that u t i l i z e d m u l t i n o m i a l l o g i t models to estimate an automobile demand f u n c t i o n suggests that income, household composition, p r i c e of g a s o l i n e , education, c e n t r a l c i t y r e s i d e n c y , and 20 employment s t a t u s are important determinants of demand. In s e c t i o n 2.5 the review of e i g h t marketing s t u d i e s that use a n a l y s i s of v a r i a n c e or d i s c r i m i n a n t a n a l y s i s to provide i n s i g h t s about automobile ownership suggests that a t t i t u d e s and p e r s o n a l i t y f a c t o r s p rovide an a d d i t i o n a l understanding of the f a c t o r s i n f l u e n c i n g car ownership. For i n s t a n c e , ownership of a p a r t i c u l a r car may i n d i c a t e p r e f e r e n c e s in terms of s i z e , "image" and brand. However, the f i n d i n g s from these s t u d i e s appear weaker than the f i n d i n g s from econometric s t u d i e s i n terms of both c o m p a r a b i l i t y of r e s u l t s and s t a t i s t i c a l r e l i a b i l i t y . In s e c t i o n 2.6 a conceptual model i s presented to i n t e g r a t e both groups of s t u d i e s . The model argues that c h o i c e s l e a d i n g to automobile ownership are determined by s i t u a t i o n a l f a c t o r s , market r e l a t e d f a c t o r s , consumer needs, and consumer c h a r a c t e r i s t i c s . F u r t h e r , i n t e r r e l a t i o n s among these f a c t o r s may a l s o generate a set of c o n s t r a i n t s , and may i n f l u e n c e p e r c e p t i o n s about automobiles. These c o n s t r a i n t s and p e r c e p t i o n s provide an a d d i t i o n a l understanding about consumer automobile ownership. In s e c t i o n 2.7 p o s s i b l e r e s e a r c h areas are i d e n t i f i e d . I t i s argued that f u t u r e r e s e a r c h may p rovide a d d i t i o n a l c l a r i f i c a t i o n of the a n a l y t i c a l i s s u e s a s s o c i a t e d with modelling ownership, and i n a d d i t i o n , may p rovide b e t t e r measurement approaches. Since past r e s e a r c h has f a i l e d to provide a c o n s i s t e n t p a t t e r n of the r e l a t i o n s h i p s between car ownership 21 and consumer a t t i t u d e s , f u t u r e r e s e a r c h should be c a r e f u l l y conducted to i s o l a t e cause and e f f e c t . I t i s noted that i t i s p o s s i b l e to examine the e f f e c t s of l i f e - s t a g e , g a s o l i n e p r i c e s , income n o n - l i n e a r i t i e s , and commuting p r a c t i c e s on automobile ownership from c u r r e n t l y a v a i l a b l e d a t a s e t s . 2.2 T h e o r e t i c a l C o n s i d e r a t i o n s The purpose of t h i s s e c t i o n i s to provide a t h e o r e t i c a l b a s i s f o r the f a c t o r s i n f l u e n c i n g automobile ownership. T h i s d i s c u s s i o n a l s o p r o v i d e s i n s i g h t s about measurement and e s t i m a t i o n i s s u e s that are c e n t r a l to the modelling problems a s s o c i a t e d with automobile ownership. One a n a l y t i c a l b a s i s from which to examine automobile ownership i s the consumption f u n c t i o n theory proposed by M o d i g l i a n i and Brumberg(1955) in t h e i r work concerning a l i f e - c y c l e consumption hypothesis, or by Friedman's(1957) permanant income model. These models argue that the goal of the i n d i v i d u a l i s to maximize the value of a l i f e t i m e u t i l i t y f u n c t i o n which i s the sum of each p e r i o d ' s u t i l i t y d i s c o u n t e d by some parameter. T h i s i m p l i e s that the consuming u n i t ' s problem i s to a l l o c a t e i t s l i f e t i m e earnings to the purchase of goods and s e r v i c e s (such as automobiles), so as to achieve maximum " s a t i s f a c t i o n " . I t was noted i n an e a r l i e r chapter that consumers use an automobile f o r a v a r i e t y of purposes. Within each purpose, 22 s a t i s f a c t i o n may be gained from s e v e r a l automobile c h a r a c t e r i s t i c s that provide convenience, comfort, speed and f l e x i b i l i t y . On the other hand, b e n e f i t s are achieved at the cos t of ownership and the a s s o c i a t e d o p e r a t i n g c o s t s . Thus, a consumer's problem with respect to automobile ownership i s to choose the number and type of c a r s to hold i n each p e r i o d , and to a l l o c a t e these c a r s among d i f f e r e n t uses such that income-cost c o n s t r a i n t s are s a t i s f i e d . A n a l y s i s of the theory. The consumption f u n c t i o n theory provides a " g l o b a l " view of consumer behaviours regarding automobiles but i t o f f e r s no i n s i g h t s about the f a c t o r s a f f e c t i n g automobile c h o i c e . To o b t a i n a l i s t of such f a c t o r s , a consumption f u n c t i o n i s analyzed i n Appendix I with the assumption that the u n d e r l y i n g u t i l i t y f u n c t i o n i s convex and twice d i f f e r e n t i a b l e . The r e s u l t s of t h i s a n a l y s i s p rovide a l i s t of f a c t o r s a f f e c t i n g car ownership, and i d e n t i f y i s s u e s a s s o c i a t e d with the measurement and e s t i m a t i o n of consumer car needs. Four major f a c t o r s appear to a f f e c t car c h o i c e . F i r s t , a consumer's expected or .actual l i f e time budget a c t s as a c o n s t r a i n t to automobile ownership and o p e r a t i o n . In consumer rese a r c h , a consumer's c u r r e n t earnings provide a proxy f o r t h i s c o n s t r a i n t . Second, over a l i f e t i m e , the number and type of automobiles owned and housing c h o i c e s are r e l a t e d . Thus, i n e m p i r i c a l r e s e a r c h when a s i n g l e demand equation i s used to e x p l a i n automobile demand, housing c h o i c e must be t r e a t e d as an 23 exogenous f a c t o r . T h i r d , commuting d i s t a n c e and/or the o p e r a t i n g c o s t a l s o a f f e c t automobile c h o i c e . F i n a l l y , i f there i s no l i m i t on borrowing or l e n d i n g , then the consumer's c h o i c e of automobiles would depend only on the cost of l e n d i n g and / or borrowing and not the p r i c e of the automobile. With a l i m i t on borrowing or l e n d i n g , however, automobile c h o i c e would depend on both the p r i c e of automobiles as well as the c o s t of borrowing and l e n d i n g . S e v e r a l i s s u e s regarding measurement and e s t i m a t i o n should a l s o be noted. F i r s t , although o l d and new c a r s may provide i d e n t i c a l t r a n s p o r t a t i o n s e r v i c e , newer c a r s may have g r e a t e r comfort, a e s t h e t i c appeal and lower v u l n e r a b i l i t y to major r e p a i r s . T h i s implies that consumers may co n s i d e r newer automobiles to be " s u p e r i o r " to o l d e r one. Thus, to provide the c a u s a l s t r u c t u r e u n d e r l y i n g consumer car ownership, car c h a r a c t e r i s t i c s must a l s o be c o n s i d e r e d . The c h a r a c t e r i s t i c s p rovide proxy measures of the b e n e f i t s sought by consumers. Second, there i s an important d i s t i n c t i o n between d e c i s i o n s about o p e r a t i n g and purchasing an automobile. While the l a t t e r i m p l i e s a . s u b s t a n t i a l change in a consumer's automobile f l e e t , the former r e f l e c t s gradual d e p r e c i a t e d change and d e r i v a t i o n of s a t i s f a c t i o n from the f l e e t . Both d e c i s i o n s are a s s o c i a t e d with ownership and each i n v o l v e s changes to the automobile f l e e t . Each d e c i s i o n r e q u i r e s a d i f f e r e n t measurement techniques i f i t i s to be s t u d i e d . 24 T h i r d , i f a demand f u n c t i o n i s to be estimated at the consumer l e v e l , then i t must be recognized t h a t the consumer's ch o i c e between v a r i o u s l e v e l s of car ownership i s not a continuous v a r i a b l e , and the a s s o c i a t e d u t i l i t y f u n c t i o n i s not convex. In other words, changes i n the l e v e l of car ownership ( i . e . , number and type of car) and the c o s t s and b e n e f i t s a s s o c i a t e d with i t , are not continuous. For example, i f the consumer owns a f u e l e f f i c i e n t Volkswagen B e e t l e and uses i t f o r commuting every day, then the o p e r a t i n g c o s t w i l l i n c r e a s e p r o p o r t i o n a l l y to the t o t a l commuting need. However, i f that consumer wishes to a c q u i r e a g r e a t e r l e v e l of s a f e t y and comfort, he or she may have to purchase another automobile at a s u b s t a n t i a l i n c r e a s e i n cost while only a c h i e v i n g a marginal in c r e a s e i n b e n e f i t s other than s a f e t y . The i m p l i c a t i o n s of t h i s a n a l y s i s suggests that the r e l a t i o n s h i p between cost and b e n e f i t may not be a continuous f u n c t i o n and hence techniques such as Ordinary Least Squares (OLS) which r e q u i r e independent e r r o r s t r u c t u r e are not a p p r o p r i a t e f o r model e s t i m a t i o n of t h i s s o r t . To summarize, automobile ownership can be c o n c e p t u a l i z e d i n terms of a l i f e time consumption f u n c t i o n , and i s r e l a t e d to housing c h o i c e , automobile o p e r a t i n g c o s t , and the cost of l e n d i n g or borrowing money. In such a theory, car c h a r a c t e r i s t i c s provide a proxy f o r the b e n e f i t s of ownership. F i n a l l y , ' f o r the e s t i m a t i o n of a consumer l e v e l demand f u n c t i o n , based on the consumption f u n c t i o n theory, OLS appears to be an 25 i n a p p r o p r i a t e technique. 2.3 Measurement and E s t i m a t i o n of Car Demand Fun c t i o n The purpose of t h i s s e c t i o n i s to d i s c u s s p o s s i b l e b e h a v i o u r a l measures r e l a t e d to automobile ownership and to compare s e v e r a l s t a t i s t i c a l techniques which are a v a i l a b l e f o r e s t i m a t i n g the r e s u l t i n g demand f u n c t i o n . Measurement problems. Consumer automobile ownership i m p l i e s three i n t e r r e l a t e d consumer a c t i o n s ; p urchasing, h o l d i n g , and changing ownership of the consuming u n i t ' s car " f l e e t " . These three behaviours c o u l d be measured on dichotomous, polytomous or i n t e r v a l s c a l e s (Table 2.1). In t h i s s u b s e c t i o n the advantages and disadvantages of v a r i o u s combinations of these p o s s i b i l i t i e s are d i s c u s s e d . A consumer's d e c i s i o n to purchase a car can be c h a r a c t e r i z e d by a simple dichotomous yes or no measure. Although such a v a r i a b l e i s simple to measure, i t ignores consumer m o t i v a t i o n s and d e c i s i o n i n t r i c a c i e s . F u r t h e r , i t p r o v i d e s no i n f o r m a t i o n about the nature of the automobile purchased or r e l a t e d v a r i a b l e s such as g a s o l i n e consumption. Since a major i n t e r e s t of t h i s r e s e a r c h i s the g a s o l i n e consumption p a t t e r n , more comprehensive measures are needed. There are s e v e r a l p o s s i b i l i t i e s . For example, automobile engine s i z e (measured e i t h e r i n number of c y l i n d e r s (Greenlees, 1980), or engine displacement 26 T a b l e 2.1 Consumer D e c i s i o n s and C h o i c e of Dependent, V a r i a b l e s Consumer S c a l e of Measurement dec i s i on D i chotomous Po1ytomous Interva1 V e h i c l e • Yes-No • Number of One or more of p u r c h a s 1 n g • Intended c y l i n d e r s fo11ow1ng -not i n t e n d e d ( G r e e n l e e s , 1 9 8 0 ) - Purchase p r i c e • S t y l e - O p e r a t i n g c o s t - subcompact - Automobile s i z e - compact p h y s i c a l volume or - s t a n d a r d passenger weight - l u x u r y - Luggage c a p a c i t y (Lave and T r a i n , 1979) • Newness - one year - two y e a r s - t h r e e y e a r s - f o u r y e a r s - f i ve o r more y e a r s • Uniqueness Veh i c1e • Yes - No • Number of Asset v a l u e of hoi d i ng c a r s automob i 1 e s ( C r a g g and F l e e t s i z e U h l e r . 1970) Passenger and • Number of baggage c a r r y i n g c y l i n d e r s capac i t y » Number and F l e e t e f f i c i e n c y type of f l e e t • Number and newness (Johnson, 1978) Change i n • Change - • S t a t u s quo. Change i n a s s e t v a l u e h o i d i ng s t a t u s quo r e p l a c e a c a r . Change In f l e e t s i z e p a t t e r n add a c a r , and Change i n baggage and se l 1 a c a r passenger c a r r y i n g ( C r a g g and capac i t y U h l e r , 1970) Change i n f l e e t • S t a t u s quo. e f f i c i e n c y add a sma11 c a r , add a l a r g e c a r . rep 1 ace a sma11 c a r e t c . ( C l a x t o n and MacD o u g a l l , 1983) 27 measured in cubic centimeters) i s one measure of the type of car purchased. A l t e r n a t i v e l y , automobile s i z e - s t y l e (Lave and T r a i n , 1979) or weight provide some info r m a t i o n about expected g a s o l i n e consumption. On the other hand, the age of the car or t o t a l k i l o m e t e r s or m i l e s t r a v e l l e d by the v e h i c l e a l s o i n d i c a t e engine wear-and-tear and provide g a s o l i n e consumption i n f o r m a t i o n . In these examples, measures based on polytomous or i n t e r v a l s c a l e s c o u l d provide reasonable i n f o r m a t i o n about the expected r a t e of g a s o l i n e usage. The consumer g a s o l i n e consumption p a t t e r n i s a c e n t r a l concern i n t h i s r e s e a r c h , however, consumers might be concerned about other b e n e f i t s from t h e i r automobile. In other words, consumers i m p l i c i t l y determine c o s t and b e n e f i t t r a d e - o f f s while g a s o l i n e consumption i s only one such c o s t . To i n v e s t i g a t e these t r a d e - o f f s , an economic measure, e i t h e r c a t e g o r i c a l (low, medium and high p r i c e ) or i n t e r v a l s c a l e ( r e a l p r i c e of car) might be r e q u i r e d . S i m i l a r l y , consumer i n t e r e s t i n purchasing unique c a r s such as sporty, or l u x u r i o u s w i l l have to be measured by s e v e r a l c a t e g o r i e s . While the o b j e c t of the purchase d e c i s i o n i s o f t e n a s i n g l e c a r , the d e c i s i o n i s i n f l u e n c e d by the consumer's t o t a l h o l d i n g of automobiles. Thus, a simple measure to i n d i c a t e t h i s i n f l u e n c e c o u l d be whether a consumer c u r r e n t l y owns any c a r ( s ) (yes or no), or a polytomous measure c o u l d be based on the number of c a r s owned [zero, one, two and three and more, (Johnson, 1978) and ( T r a i n , 1980)]. Although both of these 28 measures are easy to o b t a i n , they must be based on a s p e c i f i e d d e c i s i o n u n i t such as, a f a m i l y or a household. F u r t h e r , they may be s p e c i f i e d so that automobile c h a r a c t e r i s t i c s (weight, s i z e , and p r i c e etc.) are summarized across the f a m i l y " f l e e t " . T h i s i m p l i e s averaging or aggregating i n f o r m a t i o n , and p o s s i b l e l o s s of i n f o r m a t i o n for e m p i r i c a l r e s e a r c h . S i m i l a r l y , a change i n the composition of the consuming u n i t ' s " f l e e t " i m p l i e s a change of one or more of the cars making up the car f l e e t . Thus, a consumer purchase can be measured by the r e s u l t i n g change i n f l e e t s i z e , t o t a l number of c y l i n d e r s or t o t a l f l e e t weight. Such a change i m p l i e s a l o n g i t u d i n a l measurement and t h i s i n t r o d u c e s time r e l a t e d measurement problems. Some of these problems w i l l be r e l a t e d to such t h i n g s as sample s i z e composition, t e c h n o l o g i c a l change, and the l i k e . L i t e r a t u r e examining the change i n consumer f l e e t ownership i s l i m i t e d (Claxton and McDougall, 1983; Cragg and Uhler, 1970), and as a r e s u l t , l i t t l e i s known about these measurement problems. To summarize, car ownership r e s u l t s from purchasing, h o l d i n g and changing the composition of a consuming u n i t ' s car f l e e t . These behaviours or a c t i o n s can be measured along dichotomous, polytomous and i n t e r v a l s c a l e s , and have c l e a r i m p l i c a t i o n s f o r g a s o l i n e consuption, which i s a major i n t e r e s t i n t h i s study. Measures of automobile purchase i s complicated by (1) the need f o r a simple yet complete i n d i c a t i o n of the type of car bought, (2) the need to have accurate summary data about 29 other c a r s i n the family f l e e t , and (3) the need to r e f l e c t the dynamics of v e h i c l e ownwership by measuring l o n g i t u d i n a l changes i n t h a t v a r i a b l e . S t a t i s t i c a l i s s u e s . To understand the problems a s s o c i a t e d with the e s t i m a t i o n of automobile demand f u n c t i o n s of the types d i s c u s s e d here, c o n s i d e r a economy where the household has only two c h o i c e s ; e i t h e r to own a car or not to own a c a r . F u r t h e r , assume that a l l cars being c o n s i d e r e d f o r ownership are s i m i l a r in every r e s p e c t . F i n a l l y , assume that the only d i f f e r e n t i a t i n g c h a r a c t e r between each household i s t h e i r incomes. The e s t i m a t i o n problem i s to c o n s t r u c t a f u n c t i o n which w i l l d e s c r i b e owners and non-owners or a l t e r n a t i v e l y to d i s c r i m i n a t e between these two groups. One of the simplest approaches to t h i s problem i s to t r e a t ownership as a dependent v a r i a b l e c o n s i s t i n g of two l e v e l s (0 and 1) and use the o r d i n a r y l e a s t squares procedure to estimate the r e l a t i o n s h i p between ownership and income. The estimated c o e f f i c i e n t , i f s i g n i f i c a n t l y d i f f e r e n t from zero and p o s i t i v e in s i g n , w i l l i n d i c a t e that household income p o s i t i v e l y a f f e c t s ownership. Although t h i s simple approach might work i n some cases, there i s no guarantee that the p r e d i c t e d value of the dependent v a r i a b l e w i l l be i n the range of zero and one (Hensher and Johnson, 1981, pp. 163-170). At the same time, the e r r o r term a s s o c i a t e d with each o b s e r v a t i o n may not be independent as r e q u i r e d f o r OLS e s t i m a t i o n . One a l t e r n a t i v e to OLS which i s o f t e n proposed i s to impose r e s t r i c t i o n s on the p r e d i c t e d values 30 of the dependent v a r i a b l e and use weighted l e a s t squares procedures (Goldberger, 1964). However, comparative r e s u l t s r e p o r t e d by Dagenais (1975) suggest that the estimated parameters from t h i s procedure tend to be b i a s e d . As a r e s u l t t h i s a l t e r n a t i v e was not pursued in t h i s t h e s i s . The second approach would be to apply a n a l y s i s of v a r i a n c e techniques to the problem. T h i s r e q u i r e s t r e a t i n g income as a random v a r i a t e and then comparing the mean income of owners and non-owners. To t e s t f o r a s i g n i f i c a n t d i f f e r e n c e between the two groups, the t - t e s t c o u l d be a p p l i e d . If m u l t i p l e v a r i a t e s are i n v o l v e d , m u l t i - v a r i a t e a n a l y s i s of v a r i a n c e or d i s c r i m i n a n t a n a l y s i s c o u l d be used (Morrison, 1976, pp. 230-248). For example, a d i s c r i m i n a n t f u n c t i o n would provide estimates of the chance of m i s - c l a s s i f y i n g an i n d i v i d u a l household with one or no c a r . The s o - c a l l e d "confusion matrix" and the relevance of v a r i a t e s i n the d i s c r i m i n a n t f u n c t i o n q u a l i t a t i v e l y t e s t the e f f e c t of a v a r i a b l e . In t h i s f o r m u l a t i o n d i s c r i m i n a n t f u n c t i o n can be i n t e r p r e t e d as a demand f u n c t i o n . The d i f f i c u l t y with t h i s technique i s that i t t r e a t s income and other household c h a r a c t e r i s t i c s as random v a r i a t e s and p r o v i d e s estimates of the chance of m i s - c l a s s i f y i n g a household (McFadden, 1976). T h i s d i f f i c u l t y , however, i s of a conceptual nature, and t e c h n i c a l i s s u e s are l e s s severe. The t h i r d a l t e r n a t i v e to the problem would be to formulate the demand as a d i s c r e t e c h o i c e model as suggested by McFadden (1974) and Hensher and Johnson (1981). The b a s i c aim of these 31 models i s to p r e d i c t the odds of an i n d i v i d u a l choosing no car over one car based on f a m i l y income. To accomplish t h i s , the parameters of a l o g i s t i c f u n c t i o n are estimated using a maximum l i k e l i h o o d procedure. Thus, f o r the above example the demand f u n c t i o n would be (1) l o g C p ^ p o ) = 0o + j3, (Income) + e with p 0 and p, are observed p r o p o r t i o n s of no and one car r e s p e c t i v e l y and /30 and /?, are the unknown parameters to be estimated. Note that i n (1), e i s not normally d i s t r i b u t e d but comes from a log-normal d i s t r i b u t i o n . F u r t h e r , e x p l i c i t l y e n t e r i n g the p r e d i c t e d values of the dependent v a r i a b l e (which are always w i t h i n the range of zero to one) in the e s t i m a t i n g equation, ensures that p 0 + P i = 1. F i n a l l y , j3, measures the impact of income on owning an automobile. Although i n the above example, income was the only independent v a r i a b l e , the b a s i c s t r u c t u r e of the problem fo r m u l a t i o n i s not damaged when m u l t i p l e independent v a r i a b l e s are i n t r o d u c e d . S i m i l a r l y , i f m u l t i p l e choice c a t e g o r i e s (e.g. no c a r , one small c a r , one l a r g e c a r , etc.) are p r e s e n t , the b a s i c s t r u c t u r e remains the same. However, for k c a t e g o r i e s , the (k-1) odds r a t i o s w i l l be estimated simultaneously to o b t a i n s t a t i s t i c a l l y e f f i c i e n t and unbiased est i m a t e s . While hypothesis t e s t i n g i s s i m i l a r to that of usual r e g r e s s i o n a n a l y s i s procedures, the measure for " g o o d n e s s - o f - f i t " i s d i f f e r e n t . An i n f o r m a l approach to measure 32 " f i t " i s to use the "confusion matrix", as i n d i s c r i m i n a n t a n a l y s i s . In the matrix, the r a t i o of the t o t a l number of cases along the d i a g o n a l to sample s i z e serves to i n d i c a t e the e f f e c t i v e n e s s of the model. A more formal measure i s analogous to a goodness of f i t measure in r e g r e s s i o n a n a l y s i s , and hence i s c a l l e d "pseudo R2". I t i s d e f i n e d as one minus the r a t i o of the l o g - l i k e l i h o o d f u n c t i o n evaluated i n the presence of a l l c o e f f i c i e n t s and the same f u n c t i o n evaluated in the absence of a l l c o e f f i c i e n t s . The measure i s s i m i l a r to R 2 and ranges between zero and one and i s always p o s t i v e . 2.4 Econometric Examples of Automobile Demand Functions In the l a s t 30 years there have been s e v e r a l s t u d i e s that attempted to c o n s t r u c t a demand f u n c t i o n f o r automobiles. These s t u d i e s were of two d i s t i n c t types. The f i r s t group of s t u d i e s employed a multinominal l o g i t - t y p e f u n c t i o n with p r e d i c t o r s r e s t r i c t e d to demographic measures. The second group of s t u d i e s employed a n a l y s i s of v a r i a n c e or d i s c r i m i n a n t a n a l y s i s procedure with s t u d y - s p e c i f i c p s y c h o l o g i c a l or s p e c i f i c demographic v a r i a b l e s (e.g. Blacks vs. Whites) as independent v a r i a b l e s . T h i s s e c t i o n summarizes the f i r s t group of these s t u d i e s to provide a m e t h o d o l o g i c a l overview. To a i d the comparison, s i m i l a r dependent v a r i a b l e models are grouped f o r a more thorough examination. The f i r s t group of models examines the number of c a r s owned. The second group focuses on the purchase 33 d e c i s i o n . In the f i n a l group, other r e l e v a n t models are d i s c u s s e d and compared. For c o n s i s t e n c y of comparison ac r o s s the s t u d i e s , the d i s c u s s i o n of a l l three groups emphasizes r e p o r t e d c o e f f i c i e n t s that are s i g n i f i c a n t at p<0«lO. It i s noteworthy that the f i v e s t u d i e s d i s c u s s e d i n t h i s s e c t i o n r e p o r t e d an average "pseudo-R 2" of 0-30. T h i s i m p l i e s that the u n d e r l y i n g demand f u n c t i o n s p r o v i d e a weak but perhaps reasonable r e p r e s e n t a t i o n of the data s e t s . Cragg and Uhler (1970) attempted to c o n s t r u c t a demand f u n c t i o n on the b a s i s of a u t i l i t y maximization model. The data c o n s i s t e d of a m u l t i - p e r i o d c r o s s - s e c t i o n sample of 986 spending u n i t s , and was c o l l e c t e d i n 1960, 1961 and 1962 by the Survey Research Center of the U n i v e r s i t y of Michigan. Two s e t s of d e c i s i o n s were modelled. The f i r s t set c o n s i d e r e d four p o s s i b i l i t i e s r e garding an automobile ownership change (no change, s e l l a c a r , r e p l a c e a c a r , and purchase an a d d i t i o n a l c a r ) . The second set analyzed four p o s s i b i l i t i e s of car h o l d i n g s (no c a r , one c a r , two c a r s , and three or more c a r s ) . Johnson (1978) i n v e s t i g a t e d the e f f e c t of f a m i l y c h a r a c t e r i s t i c s on the demand f o r both new and used automobiles. Data from c r o s s - s e c t i o n a l surveys of consumer f i n a n c e s , conducted by the Survey Research Center of the U n i v e r s i t y of Michigan, were used to c o n s t r u c t these models. Based on n a t i o n a l samples of 2576 f a m i l i e s i n 1970, and 2860 i n 1959, the e f f e c t of income, f a m i l y s i z e and the age of the f a m i l y head were r e l a t e d to the number of c a r s owned and the age of the c a r s 34 owned. Greenlees (1980) examined the impact of g a s o l i n e p r i c e on the mix of new automobile purchases. His sample c o n s i s t e d of 1257 new c a r s purchased between January 197 1 and March 1974. Ga s o l i n e p r i c e s were obtained f o r 21 c i t i e s i n the US. A model was c o n s t r u c t e d to p r e d i c t the p r o b a b i l i t y of four or s i x c y l i n d e r c a r s being purchased. The base case was e i g h t c y l i n d e r c a r s . Lave and T r a i n (1979) c o n s t r u c t e d a disaggregate model of household car buying d e c i s i o n s . They c o l l e c t e d a s t r a t i f i e d random sample of 541 new car buyers from seven US c i t i e s in 1976. A model was c o n s t r u c t e d to p r e d i c t the p r o b a b i l i t y that households would buy one of ten auto types with the 10 c a t e g o r i e s being based on car s i z e and p r i c e . T r a i n (1980) argued that there i s a s i m u l t a n e i t y between automobile ownership and work t r a v e l mode c h o i c e s . His study estimated model c o e f f i c i e n t s to account f o r t h i s s i m u l t a n e i t y . A s t r a t i f i e d sample f o r 656 workers from the San F r a n s i s c o Bay area was c o l l e c t e d on the b a s i s of p r o x i m i t y to BART (Rapid T r a n s i t ) i n 1975. J o i n t multinominal p r o b a b i l i t i e s (mode choice and ownership) were estimated. Ownership models. Among the above s t u d i e s , there- were s i x models c o n s t r u c t e d for the number of automobiles owned (Table 2.2). Parameters f o r f i v e of these s i x models were estimated using n a t i o n a l samples [Cragg and Uhler - sample years 1960, 3E> T a b l e 2.2 Summary of S t u d i e s P r e d i c t i n g Number of Cars Owned Independent Study Measure Number of c a r s J Number of c a r s and type f a c t o r no vs . no vs . no vs . no vs. no vs . no vs . no vs . one two one one two one two used new used used. new one used Cragg Real income ++, ++, ++, ++, and - 1i near ++ ++ Uhl er - Q u a d r a t i c --, -- , -- . (1970) ++ + + WEALTH Tr a i n - 1i near + + + (1980) Johnson Logar i thm ++, ++ ++ , ++ ++ ++ ++ ++ ++ ( 1978) of income Cragg Number of ++ , ++, ++, ++, and a d u l t s ++ ++ U h l e r Number of ns, ns, ns, ns, ch i 1 d r e n ns ns FAMILY T r a i n Fami1y ++ ++ COMPOSITION s i ze % of ++ ++ a d u l t s i n fam i1y Johnson 1ogor i thm ++, ++ ++ , ++ ++ ns ++ + + + + of fam i1y s i ze Cragg Age of --, ns, * . ns, FAMILY and fam i1y -- ns LIFE-CYCLE Uhl e r head STAGE Johnson logar1thm -- . ns, ns -- -- ns ns ns of age of fam i1y head WORK Cragg More than ns, ns, * , ns RELATED and one wage ns ns FACTORS Uhl er e a r n e r s Cragg In c i t y -- . -- • and c e n t r e -- --HOUSE U h l e r LOCATION T r a i n In c i t y -- --c e n t r e COST T r a i n % of -- --FACTOR i ncome spent on c a r expenses TRANSIT • T r a i n A c c e s s to + + ACCESS non-work a r e a s by A t r a n s i t ++ or -- s i g n i f i c a n t at p + or - s i g n i f i c a n t at p < 0.01 and s i g n i n d i c a t e s d i r e c t i o n of e f f e c t < 0.05 and s i g n i n d i c a t e s d i r e c t i o n of e f f e c t s i g n i f i c a n t at p < 0.10 not s i g n i f i c a n t at p < 0.10 36 1961, 1962; Johnson, sample years 1959 and 1970] and one model used a San F r a n s i s c o Bay area sample [year 1975]. A l l s i x of these models i n c l u d e d a separate equation f o r "three or more" car owners. The sample s i z e s f o r t h i s separate equation ranged between 4 and 70 ( T r a i n ' s study had a sample s i z e of 70). Hence, the d i s c u s s i o n here i s con c e n t r a t e d on consumer choice between no, one or two c a r s . F u r t h e r , because the v a r i a b l e s were o p e r a t i o n a l i z e d d i f f e r e n t l y across these s t u d i e s , the c o e f f i c i e n t s can not be d i r e c t l y compared. However, comparisons of s i g n and s t a t i s t i c a l s i g n i f i c a n c e provide some i n d i c a t i o n of the c o n s i s t e n c y between the v a r i o u s r e l a t i o n s h i p s . A l l of the models i n c l u d e d income and found p o s i t i v e e f f e c t s ; higher income earners tended to own more c a r s . An examination of the s i g n i f i c a n c e a s s o c i a t e d with the income c o e f f i c i e n t s i n d i c a t e d that the odds r a t i o between one and no car owners was s i g n i f i c a n t at p<0-0l i n four s t u d i e s and at p<0«05 f o r the other two. The odds r a t i o between two and no car was s i g n i f i c a n t at p<0-01 in a l l s i x s t u d i e s . T h i s i m p l i e s that income can be used to d i s t i n g u i s h between two and no car owners w e l l , but r e s u l t s are somewhat weaker f o r one and no car owners. Cragg and Uhler used q u a d r a t i c terms i n a l l three s t u d i e s . Although a l l c o e f f i c i e n t s a s s o c i a t e d with an odds r a t i o of one and no car and two and no car were s i g n i f i c a n t at p^O-01, the si g n s a s s o c i a t e d with 1960 and 1961 data were negative while the si g n f o r 1962 was p o s i t i v e . Where the signs were found to be negative f o r the q u a d r a t i c term and a l a r g e r p o s i t i v e 37 c o e f f i c i e n t i s a s s o c i a t e d with the l i n e a r income term, then income and number of c a r s owned have a convex r e l a t i o n s h i p . On the other hand, where the q u a d r a t i c term had a p o s i t i v e s i g n , income and the number of c a r s owned bear a concave r e l a t i o n s h i p . As a r e s u l t , the e m p i r i c a l evidence p r o v i d e s c o n f l i c t i n g r e s u l t s about the shape of the r e l a t i o n s h i p . Both curves (concave or convex) g e n e r a l l y imply, however, that beyond some l e v e l of income, marginal i n c r e a s e s do not a f f e c t the number of c a r s owned. Age of the household head was used i n f i v e of the s i x models. Equations a s s o c i a t e d with two car owners d i d not produce s i g n i f i c a n t r e s u l t s i n any of the f i v e cases. However, equations a s s o c i a t e d with one car owners d i d show s i g n i f i c a n t negative c o e f f i c i e n t s (p^0«05) in a l l but one equation (Cragg and Uhler, 1961 d a t a ) . T h i s suggests that age does not d i s c r i m i n a t e between two and no car owners. However, age was i n v e r s e l y r e l a t e d to the p r o b a b i l i t y of a spending u n i t owning only one c a r . There were four d i f f e r e n t approaches to d e f i n i n g household composition. Cragg and Uhler c o n s i d e r e d two approaches, number of a d u l t s and number of c h i l d r e n ; Johnson used f a m i l y s i z e , and T r a i n used f a m i l y s i z e and the p r o p o r t i o n of a d u l t s i n a household. A l l d e f i n i t i o n s of household composition produced s i g n i f i c a n t (p<0«0l) p o s i t i v e c o e f f i c i e n t s a c r o s s a l l equations with one e x c e p t i o n . The exception was the number of c h i l d r e n i n a household which was i n s i g n i f i c a n t i n a l l cases. T h i s suggests 38 that f a m i l i e s with more a d u l t s tend to own more c a r s . Four s t u d i e s found that the p r o x i m i t y to the c e n t r a l business d i s t r i c t had a s i g n i f i c a n t (p<0-05) negative e f f e c t on ownership l e v e l . One study ( T r a i n , 1980) found that a c c e s s i b i l i t y to non-work d e s t i n a t i o n s by p u b l i c t r a n s i t , i n c r e a s e d the p r o b a b i l i t y of owning no car (p<0«l0). The same study a l s o found s i m u l t a n e i t y between w o r k - t r i p need and automobile ownership, suggesting that those who commute long d i s t a n c e s to work tend to have a higher number of c a r s than those who commute short d i s t a n c e s . Cragg and Uhler attempted to t e s t a s i m i l a r notion by u s i n g a v a r i a b l e denoting whether a household had more than one person employed. The r e s u l t s i n d i c a t e d that f a m i l y employment s t a t u s d i d not d i s c r i m i n a t e between no car and m u l t i p l e car owners. T h i s i m p l i e s that work t r i p d i s t a n c e and work s t a t u s may be needed to understand ownership. Johnson's study a l s o focused on s i x d i s c r e t e c h o i c e s ; no, one new, one o l d , two o l d , one o l d and one new and two new c a r ( s ) . A 1970 sample was used to estimate the p r o b a b i l i t i e s a s s o c i a t e d with a family choosing one of the s i x a l t e r n a t i v e s . The independent v a r i a b l e s were the l o g a r i t h m of household income, age of household head, and f a m i l y s i z e . The study concluded t h a t , h o l d i n g the number of cars c o n s t a n t , i n c r e a s e s in income p o s i t i v e l y a f f e c t e d the p r o b a b i l i t i e s a s s o c i a t e d with owning newer c a r s (p<0«0l). S i m i l a r l y , h o l d i n g the age of cars constant, i n c r e a s e d income r e s u l t e d i n an i n c r e a s e i n the 39 p r o b a b i l i t y of a f a m i l y owning more c a r s (p<0«0l). On the other hand, h o l d i n g the number of c a r s constant, i n c r e a s e d f a m i l y s i z e p o s i t i v e l y i n f l u e n c e d the chances f o r used car ownership (p<0«0l). S i m i l a r l y , i t was found that the e f f e c t of age of the household head was s i g n i f i c a n t only f o r one car owners. Thus, i n c r e a s e d age of the household head decreased chances of owning an automobile or owning a newer automobile. To summarize the f i n d i n g s r e g a r d i n g demographics and car ownership p a t t e r n s , income and f a m i l y s i z e i n d i c a t e d a c o n s i s t e n t p o s i t i v e r e l a t i o n s h i p while the f i n d i n g s a s s o c i a t e d with the age of the f a m i l y head were i n c o n s i s t e n t . F u r t h e r , employment s t a t u s and w o r k - t r i p need both suggested a p o s s i b l e p o s i t i v e r e l a t i o n s h i p to the s i z e of f a m i l y car h o l d i n g s . F i n a l l y , income, fami l y s i z e and age of the f a m i l y head a f f e c t e d the choice of new and used c a r s d i f f e r e n t l y . Purchase d e c i s i o n models. There were two s t u d i e s that focused on the purchase of new automobiles(Table 2.3). Greenlees'(1980) study focused on the c h o i c e of engine s i z e ( f o u r , s i x and e i g h t c y l i n d e r s ) and the Lave and T r a i n (1979) study examined the c h o i c e of. automobile type (subcompact, compact, e t c . ) . Greenlees study concluded that i n c r e a s e d r e a l g a s o l i n e p r i c e s r a i s e d the a t t r a c t i v e n e s s of four c y l i n d e r c a r s r e l a t i v e to s i x c y l i n d e r c a r s , and both f o u r s and s i x e s r e l a t i v e to e i g h t s . T h i s study i n d i c a t e d that i f r e a l g a s o l i n e p r i c e s i n c r e a s e d by three cents per g a l l o n (one standard d e v i a t i o n ) , the market share of e i g h t c y l i n d e r c a r s d e c l i n e d by about 20%. However, T a b l e 2.3 Summary of S t u d i e s P r e d i c t i n g A u t o m o b i l e Purchase C h o i c e 40 Independent Measure Consumer c h o i c e s Consumer c h o i c e s f a c t o r s used examined by examined by G r e e n l e e s ( 1 9 8 0 ) Cragg and U h l e r ( 1 9 7 0 ) P u r c h a s e of V4 v s . V8 V6 v s . V8 S e l l a Rep1 ace P u r c h a s e c y 1 i nder c y l i n d e p c a r a c a r a c a r c a r c a r v e r s u s no change Real income ns. -- ns, ns" * , - ++ - 1i near - q u a d r a t i c ns, * ns, ns * _ WEALTH P r e v i o u s y e a r s 1ncome ns. * ns, ns ns, ns T o t a l l i q u i d a s s e t *. ns ns. ns ns , + Fami1y s i z e ns * FAMILY If female head ns ++ COMPOSITON ho u s e h o l d Number of a d u l t s ns, ns ns. + * , + + Number of ns. ns ns. ns * , ns ch i 1 d r e n Age of fam i1y -- -- ns, ns ns. - ns, * head FAMILY F a m i l y w1 t h ns ns LIFE-CYCLE c h i 1 d r e n STAGE Co u p l e s ns WORK RELATED Two or more * + ns, ns ns. - ns, * FACTORS employed Three or more + employed OTHER E d u c a t i on + * DEMOGRAPHIC Race - i f non-white f a m i l y * * L i v e i n c i t y ns ns, ns +, ns * -c e n t r e HOUSE L i v e i n r u r a l ns ns LOCATION a r e a s L1ve i n U.S. i n - N o r t h - -- -- South - -'- West ' ++ ns COST FACTORS Real p r i c e of + + + g a s o l i ne CAR FLEET Own more than ++ + ++, + ns, ns ns, ns DESCRIPTION two c a r s V a l u e of c a r ns, ++ ns, ns ns, ns c a r f l e e t i n $ Age of c a r f l e e t ns. ++ ns, + * , ns ++ o r -- s i g n i f i c a n t at p < 0.01 and s i g n i n d i c a t e s d i r e c t i o n of e f f e c t . + o r - s i g n i f i c a n t at p < 0.05 and s i g n i n d i c a t e s d i r e c t i o n of e f f e c t . * s i g n i f i c a n t at p < 0.10 ns not s i g n i f i c a n t a t p < 0.101 41 i n c r e a s e d r e a l income i n c r e a s e d the chances of buying ei g h t c y l i n d e r c a r s . The s t a n d a r d i z e d c o e f f i c i e n t s suggested that g a s o l i n e p r i c e had a stronger e f f e c t on reducing the market share of e i g h t c y l i n d e r c a r s than income had on i n c r e a s i n g t h e i r ( e i g h t c y l i n d e r c a r s ) market share. Greenlees a l s o found that i n c r e a s e d age of the f a m i l y head decreased the p r o b a b i l i t y of buying four c y l i n d e r c a r s , and i n c r e a s e d the p r o b a b i l i t y of buying e i g h t s . F u r t h e r , demand for s i x c y l i n d e r e d cars d e c l i n e d modestly with changes in the age of the f a m i l y head. T h i s study a l s o concluded that l a r g e r f a m i l i e s tend to purchase e i g h t c y l i n d e r or l a r g e r c a r s . However, fami l y s i z e c o e f f i c i e n t s were s i g n i f i c a n t only at P<0»10 and a l l other f i n d i n g s were s i g n i f i c a n t at p<0«0l. Greenlees' study a l s o p r o v i d e s s i g n i f i c a n t f i n d i n g s regarding female head of household, non-white household head, education and r e g i o n a l d i f f e r e n c e s and m u l t i p l e car owner's new car purchases, without supp o r t i n g t h e o r e t i c a l reasons. Greenlees concluded that households with a female head tended to purchase s i x c y l i n d e r c a r s and households with non-white heads p r e f e r e d e i g h t or s i x c y l i n d e r c a r s . On the other hand, in c r e a s e d education appeared to f o s t e r the purchase of smaller c a r s while f a m i l i e s with multi-incomes and those with m u l t i p l e autos tended to purchase e i t h e r s i x or four c y l i n d e r c a r s . I t was a l s o noted that couples ( f a m i l i e s with husband and wife only) tended to opt f o r oix and e i g h t c y l i n d e r c a r s . 42 To study the s t a b i l i t y of the estimated c o e f f i c i e n t s , Greenlees estimated c o e f f i c i e n t s f o r pre- and post energy c r i s i s p e r i o d s . Although a c t u a l values of the c o e f f i c i e n t s showed some d i f f e r e n c e s , the d i r e c t i o n and s i z e of the c o e f f i c i e n t s with res p e c t to g a s o l i n e p r i c e s remained s i m i l a r . I t has been noted that new auto purchases i n v o l v e t r a d e - i n s and as a r e s u l t f u t u r e car purchases depend on the type of t r a d e - i n c a r . To t e s t whether t h i s a f f e c t e d the estimated c o e f f i c i e n t s , an a l t e r n a t i v e model was c o n s t r u c t e d that i n c l u d e d t r a d e - i n v e h i c l e c h a r a c t e r i s t i c s . The r e s u l t s i n d i c a t e d that although there was some s i z e - l o y a l i t y (owners of e i g h t c y l i n d e r c a r s tended to buy newer e i g h t c y l i n d e r c a r s ) , the estimated c o e f f i c i e n t s showed remarkable s i m i l a r i t y both i n terms of s i z e as w e l l as d i r e c t i o n . The Lave and T r a i n (1979) study c l a s s i f i e d a l l automobiles purchased i n t o 10 c a t e g o r i e s such that i n c r e a s e d category order g e n e r a l l y r e f l e c t e d monotonic i n c r e a s e in automobile s i z e (Subsubcompact, Sports c a r s , Subcompact-A, Subcompact-B, Compact-A, Compact-B, Intermediate, Standard-A, Standard-B, and Luxury) a major d i f f e r e n c e between t h i s study and the others was i t s i n c l u s i o n of m u l t i p l e automobile c h a r a c t e r i s t i c s (e.g. weight, performance, number of seats) to p r e d i c t c h o i c e . I t was acknowledged by Lave and' T r a i n that automobile c h a r a c t e r i s t i c s measured o b j e c t i v e l y do not vary across households, hence p e r c e i v e d automobile c h a r a c t e r i s t i c s must be entered to e x p l a i n the p r o b a b i l i t y of s p e c i f i c c h o i c e . For example, a l l households 43 wishing to purchase a sub-subcompact car (such as a VW B e e t l e ) would face i d e n t i c a l MPG r a t i n g , weight, and s e a t i n g c a p a c i t y . As a r e s u l t the v a r i a n c e necessary to estimate the c o e f f i c i e n t s of the demand equation would be absent. However, i f i t was assumed that p e r c e i v e d a t t r i b u t e s r e s u l t from the i n t e r a c t i o n of household c h a r a c t e r i s t i c s and car a t t r i b u t e s , then the necessary v a r i a n c e would be i n t r o d u c e d . For example, p e r c e i v e d cost c o u l d be approximated by the r a t i o between automobile o p e r a t i n g c o s t and income. S i m i l a r l y a measure of comfort c o u l d be c h a r a c t e r i z e d by the r a t i o of f a m i l y s i z e and automobile s i z e or the number of seats i n a l l automobiles. The Lave and T r a i n study used these types of i n t e r a c t i o n terms between household and automobile a t t r i b u t e s to p r e d i c t s p e c i f i c auto type c h o i c e s . The study f i n d i n g s suggested, that as the purchase p r i c e of c a r s i n c r e a s e d , the p r o b a b i l i t y of buying l a r g e c a r s decreased. Whereas, i n c r e a s e d income i n c r e a s e d the p r o b a b i l i t y of choosing an expensive, u s u a l l y f u e l - i n e f f i c i e n t automobile. F u r t h e r , an i n c r e a s e i n the p r i c e of g a s o l i n e i n c r e a s e d the p r o b a b i l i t y of choosing s m a l l e r , f u e l - e f f i c i e n t automobiles. S i m i l a r l y , i n c r e a s e i n income was a s s o c i a t e d with households owning autos f o r s p e c i f i c purposes, or to purchase "expensive" c a r s . Although the Lave and T r a i n study confirmed both of these e f f e c t s , the estimated c o e f f i c i e n t s suggested that the purchase of expensive c a r s was l e s s frequent than the purchase of s p e c i f i c purpose c a r s . 44 The Lave and T r a i n study a l s o confirmed that l a r g e r households tended to p r e f e r automobiles with l a r g e r s e a t i n g c a p a c i t i e s w i t h i n the s p e c i f i c auto-type category. For example, i f a c h o i c e was between a mid-size Ford and a mid-size Ford wagon, then the households with l a r g e r f a m i l i e s p r e f e r r e d the wagon. F u r t h e r , i t was a l s o confirmed that households with a g r e a t e r demand for t r a v e l tended to purchase l a r g e r automobiles. However, these c o e f f i c i e n t s were only s i g n i f i c a n t at p<0«l0. The i n t e r a c t i o n between weight and education suggested that with other t h i n g s equal, households with l i t t l e education p l a c e high value on " b i g c a r s " . In c o n t r a s t , the most educated group appeared to p r e f e r compact or subcompact c a r s . S i m i l a r l y the i n t e r a c t i o n between car weight and the age of the household head impli e d t h a t an i n c r e a s e in car weight had a stronger p o s i t i v e i n f l u e n c e on the purchase d e c i s i o n s made by the o l d e r heads of households than for the younger heads. On the other hand, i t was noted that younger purchasers were more i n f l u e n c e d by an i n c r e a s e i n performance ( a c c e l e r a t i o n ) than were o l d e r purchasers. The Lave and T r a i n problem c o n c e p t u a l i z a t i o n was i n n o v a t i v e and t h e i r modelling technique provided some b a s i c i n s i g h t s about the i n f l u e n c e of l i f e - s t y l e on the consumer d e c i s i o n p r o c e s s . However, t h e i r r e s u l t s were somewhat weak s i n c e only s i x out of 15 estimated c o e f f i c i e n t s were s i g n i f i c a n t at p^0*05. In c o n t r a s t , the Greenlees study rep o r t e d 20 out of 34 c o e f f i c i e n t s s i g n i f i c a n t at the same l e v e l . T h i s may have been the r e s u l t of 45 Greenlees using a l a r g e r and more hetrogenous sample than that used by Lave and T r a i n . To summarize, both s t u d i e s concluded that g a s o l i n e p r i c e n e g a t i v e l y a f f e c t e d the s i z e of the car purchased. Higher household income i n c r e a s e d the chances of a household owning bigger and/or more expensive c a r s and l a r g e r f a m i l i e s tended to p r e f e r c a r s with l a r g e r s e a t i n g c a p c i t i e s . Higher education, however, was a s s o c i a t e d with the purchase of smaller c a r s . Other r e l e v a n t models. There was only one study (Cragg and Uhler, 1970) that p r e d i c t e d the p r o b a b i l i t i e s of a household s e l l i n g , r e p l a c i n g , adding to or not changing the household's car h o l d i n g s . Although they i n c l u d e d 15 v a r i a b l e s to p r e d i c t the v a r i o u s c h o i c e s only, income showed a c o n s i s t e n t p a t t e r n i n terms of the r e s u l t i n g parameter esti m a t e s . I t was observed that i n c r e a s e d income decreased the p r o b a b i l i t y of a household s e l l i n g a c a r , and at the same time in c r e a s e d the chances of a household r e p l a c i n g a car or purchasing an a d d i t i o n a l c a r . I t was a l s o confirmed that t h i s e f f e c t i s n o n - l i n e a r and p o s s i b l y quadrat i c . In a d d i t i o n , Cragg and Uhler r e p o r t e d that the number of a d u l t s i n the f a m i l y , l e v e l of l i q u i d a s s e t s (e.g., cash), and age of c a r s c u r r e n t l y h e l d , p o s i t i v e l y a f f e c t e d chances of purchasing an a d d i t i o n a l c a r . C e n t r a l c i t y r e s i d e n c y and the age of the household head, however, n e g a t i v e l y a f f e c t e d the chance of purchasing an a d d i t i o n a l c a r . I t was a l s o r e p o r t e d that a f a m i l y with an o l d e r head of household, a f a m i l y with 46 more a d u l t s , and a f a m i l y with an o l d e r car was l i k e l y to r e p l a c e a c a r . F i n a l l y , a f a m i l y with higher income, and a f a m i l y with higher l i q u i d a s s e t s was l e s s l i k e l y to s e l l a c a r . With the e x c e p t i o n mentioned above, a l l other c o e f f i c i e n t s f a i l e d to show c o n s i s t e n t p a t t e r n s of si g n s or were not s i g n i f i c a n t l y d i f f e r e n t from zero. Although t h i s d i s c u s s i o n has concentrated on s t u d i e s using formal models, there were a l s o i n f o r m a l attempts to d e s c r i b e the p a t t e r n of household automobile ownership. Dobson and Larson (1979) in an attempt to show the r e l a t i o n s h i p between socio-economic and p s y c h o l o g i c a l f a c t o r s and automobile s i z e used three d i f f e r e n t data s e t s . They found that owners of s m a l l e r , more f u e l - e f f i c i e n t , v e h i c l e s tended to be more, aware of an automobile f u e l - e f f i c i e n c y l a b e l at the time of purchase. Such awareness was a l s o p o s i t i v e l y r e l a t e d to education l e v e l . I t was a l s o found that high income households tended to buy l a r g e r automobiles, and that consumers tended to r e p l a c e automobiles with comparable s i z e d v e h i c l e s . The Lerman and Ben-Akiva (1976) study confirmed that income, household s i z e , and employment s t a t u s p o s i t i v e l y a f f e c t e d the number of c a r s owned. The study a l s o presented st r o n g evidence that the number of c a r s owned was r e l a t e d to the number of c h i l d r e n i n the household. Although Emmerson and Down (1982) c o l l e c t e d household data at Reading, England f o r the years 1962, 1971 and 1976, t h e i r sample s i z e (n=75) was small and d i d not allow them to c o n s t r u c t 47 a formal model. However, u n i v a r i a t e s t a t i s t i c s r e p o r t e d by them suggested that income and household s i z e were p o s i t i v e l y a s s o c i a t e d with the number of autos owned. To conclude, past e m p i r i c a l models have focused on (1) ownership l e v e l , (2) purchase, and (3) change in ownership d e c i s i o n s . V a r i a b l e s such as income, household s i z e , p r i c e of g a s o l i n e , education, l o c a t i o n of house and employment s t a t u s have appeared with a c o n s i s t e n t p a t t e r n of c o e f f i c i e n t s i n terms of s i g n and magnitude a c r o s s the v a r i o u s s t u d i e s d i s c u s s e d . 2.5 Marketing Examples of Automobile Ownership Stud i e s As with the s t u d i e s d i s c u s s e d i n the p r e v i o u s s e c t i o n , a l l s t u d i e s d i s c u s s e d i n t h i s s e c t i o n are summarized with respect to t h e i r key f i n d i n g s and a comparative a n a l y s i s i s presented i n three c a t e g o r i e s . The f i r s t category examines p r e f e r e n c e s regarding car s i z e , while the next category summarizes s t u d i e s concerned with the e f f e c t s of "car image" on p r e f e r e n c e s . In the f i n a l category, a review of s t u d i e s focused on brand l o y a l i t y i s r e p o r t e d . I t must be noted that most b e h a v i o u r a l r e s e a r c h d e a l i n g with energy and automobile usage has focused upon g a s o l i n e consumption (Signore and K a s s a r j i a n , 1981, R i t c h i e e t . a l . 1982, Katz, 1980), t r a v e l mode c h o i c e s ( W a l l i n and Wright, 1974, Recker and Golob, 1976 and M i t c h e l s o n and G a u t h i e r , 1979), and p s y c h o l o g i c a l d i s p o s i t i o n s . L i t t l e i n the l a s t f i v e years has 48 been r e p o r t e d about automobile ownership and a t t i t u d e s . As a r e s u l t , the m a t e r i a l reported here i s somewhat dated. However, the purpose of reviewing the work i s not focused on s p e c i f i c consumer group p r e f e r e n c e s , but on p r o v i d i n g a broader conceptual framework of automobile ownership f a c t o r s . Evans (1959), i n h i s c l a s s i c study, used the Edwards Personal P r e f e r e n c e Schedule (EPPS) to p r e d i c t ownership of Fords and C h e v r o l e t s . In 1958, he surveyed 71 Ford owners and 69 C h e v r o l e t owners from Park F o r e s t , I l l i n o i s and concluded that the p e r s o n a l i t y f a c t o r s achievement, deference, e x h i b i t i o n , autonomy, a f f i l i a t i o n , i n t r a c e p t i o n , dominance, abasement, change and aggression were able to d i s c r i m i n a t e between two brands c o r r e c t l y f o r 62.9% of h i s sample. However, i n c l u d i n g car c h a r a c t e r i s t i c s and consumer demographic data only improved the d i s c r i m i n a t i o n a b i l i t y of h i s model by 7%. T h i s , he concluded, was a weak i n d i c a t i o n that brand p r e f e r e n c e s were centered i n automobile c h a r a c t e r i s t i c s . He was a l s o able to reproduce almost i d e n t i c a l r e s u l t s with another sample of households in 1965 (Evans, 1968). w e s t f a l l (1962) attempted to show that there were fundamental p e r s o n a l i t y d i f f e r e n c e s among c o n v e r t i b l e , compact and standard car owners. He used Thurstone's Temperament Schedule which measured such temperament v a r i a b l e s as a c t i v e n e s s , v i g o u r , impulsion, dominance, s t a b i l i t y , s o c i a b i l i t y and r e f l e c t i o n . His sample c o n s i s t e d of 231 car owners from the Chicago and Los Angles areas. He concluded that standard and 49 compact car owners showed s i m i l a r p e r s o n a l i t y p r o f i l e s . C o n v e r t i b l e car owners, however, showed higher scores on s c a l e s r e l a t i n g to a c t i v e n e s s , v i g o u r , impulsiveness, dominance and soc i a b i 1 i t y . I to (1967) argued that there were s i g n i f i c a n t d i f f e r e n c e s between brand l o y a l owners and owners who switch brands. He used part of a 1963 n a t i o n a l sample of 211 Ford and 366 Chevrolet owners. A t t i t u d e s towards both c a r - t y p e s were measured by a 11-point s c a l e d e a l i n g with automobile c h a r a c t e r i s t i c s . These c h a r a c t e r i s t i c s measured automobile r e l i a b i l i t y , c o s t performance, and imagery. The study r e v e a l e d that a t t i t u d e s a f f e c t e d brand l o y a l i t y . For example, the study i n d i c a t e d that Ford owners repurchasing another Ford, and Ford owners s w i t c h i n g to C h e v r o l e t were s i g n i f i c a n t l y d i f f e r e n t i n t h e i r a t t i t u d i n a l d i s p o s i t i o n s . However, he d i d not report other d i f f e r e n c e s between Ford and Ch e v r o l e t owners. Akers (1967) compared automobiles owned by Blacks with those owned by Whites to pr o v i d e evidence that d i f f e r e n c e s i n " s o c i a l i z a t i o n " a f f e c t consumption p a t t e r n s . He used a Chicago area sample of 600 that was matched by income, education and age. He found that h o l d i n g income constant Blacks owned ca r s that were higher p r i c e d and had more c y l i n d e r s than d i d White owners. However, no d i f f e r e n c e s were found between the races i n terms of the age of the car owned, number of years the car was owned, or whether the car purchased was a new or used model. 50 The study a l s o concluded that measures of s e l f p e r c e p t i o n d i d not e x p l a i n the type of car owned. Bi r d w e l l (1968) argued that an automobile was chosen fo r i t s symbolic meaning and as a p h y s i c a l extension of p e r s o n a l i t y . He hypothesized that self-image needs were d i r e c t l y r e l a t e d to purchasing behaviour, and that there was a congruence between an automobile owner's s e l f p e r c e p t i o n and h i s or her p e r c e p t i o n of the car owned. He randomly sampled, from a county i n Texas, 100 purchasers of p r e s t i g e , mid-priced, low-priced and compact c a r s . He concluded that owners of p r e s t i g e cars were s o c i a l and s t y l i s t i c pace s e t t e r s and were economically u n c o n s t r a i n e d . Thus, these car owners had c a r s that were most congruent to t h e i r needs. At the other end of the economic la d d e r , a c t u a l i z a t i o n of wishes showed l i t t l e congruence due to economic c o n s t r a i n t . Thus, for example, i t was argued that an unemployed i n d i v i d u a l may wish to own a Porsche, but economically i t may not be p o s s i b l e . Grubb and Hupp (1968) argued that consumers of a s p e c i f i c brand of a product would have s e l f - c o n c e p t s s i m i l a r to those they a t t r i b u t e to other consumers who.buy the same brand. To t e s t t h i s h y p o t h e s i s , a sample of 36 Pontiac and 45 Volkswagen owners was c o l l e c t e d from the automobile r e g i s t r a t i o n l i s t at the U n i v e r s i t y of Nebraska. I t was noted that owners of smaller c a r s (Volkswagen) rated themselves s i g n i f i c a n t l y lower than the owners of l a r g e r c a r s (Pontiac) on such t r a i t s as s t a t u s c o n s c i o u s , f a s h i o n a b l e , adventurous, and s p o r t i n e s s . I t was 51 a l s o noted that small car owners c o n s i d e r e d themselves to be t h r i f t y and economy minded. Wiseman (1971) c o l l e c t e d from the New York region a random sample of 210 people who purchased a new automobile d u r i n g the "new model i n t r o d u c t i o n p e r i o d " f o r the year 1969. The study was designed to determine the d i f f e r e n c e s between consumers who bought "new" c a r s and those who bought " l e f t o v e r " c a r s . I t was concluded that buyers of l e f t o v e r i ntermediate s i z e c a r s were young, d e s i r e d to save by shopping around, and p r e f e r r e d b a s i c and economical automobiles. F u l l s i z e d " l e f t o v e r " models were purchased by those of middle age.and middle income with the l e a s t i n t e r e s t i n s t y l i n g . F u r t h e r , the c h a r a c t e r i s t i c s a s s o c i a t e d with purchase of l e f t o v e r f u l l s i z e c a r s were expected l e n g t h of ownership, number of c h i l d r e n , and d e s i r e f o r q u a l i t y at s u b s t a n t i a l s a v i n g . In c o n t r a s t , consumers with fewer c h i l d r e n , higher income and the l e a s t d e s i r e f o r economy purchased new intermediate s i z e d c a r s . S i m i l a r l y , the w e a l t h i e s t and o l d e s t of the sample purchased new f u l l s i z e and high p r i c e d automobiles. F i n a l l y , new f u l l s i z e , and low p r i c e d automobiles were purchased by those with the. f e e l i n g that i t was important f o r them to own a new c a r . In summary, i t was concluded that consumer shopping p a t t e r n s and expected car uses p r e d i c t e d the type of automobile purchased d u r i n g the new model t r a n s i t i o n p e r i o d . 52 Automobile s i z e p r e f e r e n c e s . There were two s t u d i e s (Akers, 1968; Wiseman, 1971) with i n f e r e n c e s regarding automobile s i z e p r e f e r e n c e s . The Akers study r e p o r t e d that Blacks tended to own c a r s with l a r g e r numbers of c y l i n d e r s (p<0«0l) than do non-Blacks. H i s f i n d i n g s a l s o support the c o n c l u s i o n that higher income earners tend to own l a r g e r c a r s . S i m i l a r l y , those who d r i v e more, tend to own l a r g e r c a r s . F u r t h e r , h o l d i n g other consumer a t t r i b u t e s constant, education was i n v e r s e l y r e l a t e d to car s i z e . Akers' study resembles i n many r e s p e c t s , s t u d i e s reported in the pr e v i o u s s e c t i o n and the c o n c l u s i o n s were s i m i l a r . However, there were two major concerns r e g a r d i n g t h i s work. Since the number of c y l i n d e r s can only take three d i f f e r e n t v a l u e s , Akers' use of OLS should be questionned (see s e c t i o n 2.3). Second, a l l of h i s f i n d i n g s were based on m u l t i p l e r e g r e s s i o n a n a l y s i s where the study obtained an R 2 of 0*08. Wiseman's study suggested that small car purchasers were g e n e r a l l y f i r s t time car buyers and young i n age. F u r t h e r , small c a r s were purchased by consumer's with an i n t e r e s t i n economy. A c l o s e r examination of these r e s u l t s a l s o r e v e a l e d that h o l d i n g image constant (image a s s o c i a t e d with l e f t o v e r or new season c a r s ) , small car owners tended to be "bargain hunters" and tended to have smal l f a m i l i e s . I n t u i t i v e l y , t h i s group seemed to comprise of young couples. 53 Image and s t a t u s r e l a t e d p r e f e r e n c e s . There were four s t u d i e s that examined the notion that the s e l f - p e r c e p t i o n s of car owners vary by type of v e h i c l e owned. For example, Grubb and Hupp argue that Volkswagen owners p e r c e i v e themselves as economy minded, p r a c t i c a l , and t h r i f t y while Pontiac owners see themselves as f a s h i o n a b l e , p l e a s u r e seekers, s t a t u s and s t y l e - c o n s c i o u s , and s p o r t y . S i m i l a r l y , W e s t f a l l r e p o r t e d that c o n v e r t i b l e owners viewed themselves as v i g o r o u s , a c t i v e , impulsive and s o c i a b l e . B i r d w e l l , and Wiseman, in s i m i l a r s t u d i e s , r e p o r t e d that there were d i f f e r e n c e s i n the purchasing behaviour between s t a t u s seekers and need s a t i s f i e r s . I t was noted in s e c t i o n 2.2 that there can be p e r c e i v e d d i f f e r e n c e s between new and o l d c a r s . F u r t h e r , "new model c a r s " at the time of i n t r o d u c t i o n , can symbolize p r e s t i g e and a consumer's preference f o r s t y l e . Wiseman not only provided support f o r t h i s n o t i o n , but a l s o concluded that consumers who purchased new models were l e s s economy minded and tended to purchase a car at two to three year i n t e r v a l s . S i m i l a r l y , B i r d w e l l a l s o r e p o r t e d that purchasers of p r e s t i g e c a r s were those with the l e a s t economic and s o c i a l c o n s t r a i n t s . However, the c h o i c e of other than p r e s t i g e automobiles appeared to depend upon the f l e x i b i l i t y and adaptiveness of s i t u a t i o n a l c o n s t r a i n t s . T h i s i m p l i e d that i n the absence of economic and s o c i a l c o n s t r a i n t s , a l l consumers may be p r e s t i g e seekers' and q u e s t i o n i n g them about t h e i r p r e f e r e n c e s alone may not i n d i c a t e purchasing or ownership 54 determinants. In summary, the past r e s e a r c h on consumer symbolic needs suggests that a m a j o r i t y of consumers a c t i v e l y a c q u i r e automobiles as s t a t u s symbols. The r e s e a r c h , however, i s l e s s than s y s t e m a t i c . For example, not a l l c a r - t y p e s nor a l l symbols have been sampled in the r e s e a r c h . F u r t h e r , d i s t i n c t i o n between a c t u a l ownership experience and symbolized ownership i s r a r e l y p r o v i d e d . Brand p r e f e r e n c e s • If s e r v i c e - l i f e expectancy and i n t e r - p u r c h a s e time a f f e c t e d brand p r e f e r e n c e s , then most durables with an average s e r v i c e l i f e - e x p e c t a n c y of 15 to 20 years (Ruff in and T i p p e t , 1975), should have l i t t l e brand l o y a l i t y . Automobile s e r v i c e l i f e - e x p e c t a n c y i s , on the average about 10 to 15 years (Greene and Chen, 1981). However, there i s an important d i s t i n c t i o n between the automobile and other consumer d u r a b l e s . While the market f o r t r a d e - i n s e x i s t s f o r a l l d u r a b l e s , t h i s type of market i s b e t t e r developed f o r the automobile than f o r other d u r a b l e s . T h i s i m p l i e s that consumers who are u n s a t i s f i e d with t h e i r automobile can e a s i l y trade i t i n the second-hand market and purchase a more p r e f e r r e d brand. T h i s i s one p o s s i b l e e x p l a n a t i o n f o r the higher than expected brand l o y a l i t y found i n the automobile market. No study has yet been able to provide c o n c l u s i v e evidence to support the r e l a t i o n s h i p between s p e c i f i c brand ownership or purchase c h o i c e s and the demographic or p s y c h o l o g i c a l c h a r a c t e r i s t i c s of owners. Evans in h i s landmark study 55 concluded that Chevrolet owners showed lower e x h i b i t i o n , g r e a t e r autonomy, g r e a t e r a f f i l i a t i o n (p<0*lO) and lower dominance (p^O'05) than Ford owners. He a l s o concluded that Chevrolet owners tended to own olde r c a r s and have smaller f a m i l i e s . K a s s a r j i a n (1971) i n h i s review of p e r s o n a l i t y r e l a t e d r e s e a r c h remarked that "to expect the i n f l u e n c e of p e r s o n a l i t y v a r i a b l e s to account f o r a l a r g e p o r t i o n of the v a r i a n c e i s most c e r t a i n l y asking too much". T h i s , of course, c o n t r a d i c t s the W e s t f a l l study where i t was re p o r t e d that there were s i g n i f i c a n t a s s o c i a t i o n s between temperament f a c t o r s and the owners of compact versus c o n v e r t i b l e car owners. However, W e s t f a l l ' s study r e p l i c a t e s Evans' f i n d i n g s of brand l o y a l i t y . On the other hand, I t o measured consumer a t t i t u d e s towards automobile r e l i a b i l i t y , c ost performance, and imagery and concluded that these a t t i t u d e s a f f e c t e d potent i a l "brand l o y a l s " and "s w i t c h e r s " between Ford and C h e v r o l e t car owners. To conclude, i t i s p o s s i b l e to observe brand l o y a l i t y i n automobile ownership, however, e m p i r i c a l r e s e a r c h has not prov i d e d c o n s i s t e n t and comparable methods of r e l a t i n g any measurement of brand l o y a l i t y to p e r s o n a l i t y f a c t o r s . 2.6 An I n t e g r a t e d Model of Automobile Ownership In an attempt to draw together the re s e a r c h d i s c u s s e d i n the p r i o r s e c t i o n s , F i g u r e 2.1 o u t l i n e s f a c t o r s i n f l u e n c i n g automobile ownership. These f a c t o r s can be c a t e g o r i z e d as three F i g u r e 2.1 I n t e g r a t e d Model of Automobile Ownersh [cost of i ,'Ownership i [•cost of I j c a p i t a l i *c. J J"Cc7st "of !Compliments [ • p r i c e of ! g a s o l i n e C-) MARKET DETERMINED FACTORS ICost "of S u b s t i t u t e • p r i c e of p u b l i c t r a n s i t A v a i l a b i l i t y of Automobile's i n Market PERCEPTION of ALTERNATIVES • p u b l i c t r a n s i t p r e f e r e n c e ( + •) Comfort and Convenience I t s i z e p r e f e r e n c e s ! i _ — i CONSUMER NEEDS from AUTOMOBILE [Travel needs • s a f e t y , . r e l i a b i l i t y • A f f e c t i v e and !Symbolic needs i ^ s t a t u s s e e k i n g SOCIAL and ECONOMIC CONSTRAINTS • r a t i o of income and t r a v e l l i n g expenses C-1 C-) AUTOMOBILE OWNERSHIP •number of aut o m o b i l e s and t h e i r c h a r a c t e r i s t i c s c-o SATISFACTION w i t h CURRENT FLEET • r a t e of c a r change j — H-"Employment j j S t a t u s i* !• number of j employed J E d u c a t i o n ! [Wealth ••annual j e a r n i n g s j-i i C + ) CONSUMER CHARACTERISTICS 1 •Family C o m p o s i t i o n ( '•number i n f a m i l y j i t i o n w — 1 IHouse Local {•distance fromf-! CBD j i ^ P e r s o n a l i t y and"! i | l A t t i t u d i n a l LP_i^pp_s_i_ t i o n | 3 58 b a s i c types - market determined f a c t o r s , consumer automobile needs, and consumer c h a r a c t e r i s t i c s . Each of these types and t h e i r i n t e r r e l a t i o n s h i p s are d i s c u s s e d below. Basic f a c t o r s . The f i r s t set of f a c t o r s that i n f l u e n c e s automobile ownership are market determined and in c l u d e four major components; c o s t of ownership, cost of compliments, cost of s u b s t i t u t e , and the a v a i l a b i l i t y of automobiles in the market p l a c e . Higher c o s t s of ownership such as the c o s t of c a p i t a l , insurance, and r e g i s t r a t i o n may i n h i b i t car ownership. Costs of compliments, however, depends upon consumer d e c i s i o n s about car usage as w e l l as g a s o l i n e p r i c e s , t o t a l d i s t a n c e t r a v e l l e d , maintenance c o s t , and rate of d e p r e c i a t i o n . Impact of these f a c t o r s on ownership may not be simple. For example, g a s o l i n e p r i c e s may i n v e r s e l y a f f e c t automobile ownership. On the other hand, the t o t a l o p e r a t i n g cost - t o t a l car usage, maintenance c o s t and r a t e of d e p r e c i a t i o n - depends upon consumer c h o i c e s about car usage. Car usage or expected car usage, however, w i l l have a p o s i t i v e impact on car ownership. S i m i l a r l y , higher c o s t and or " u n a v a i l a b i l i t y " of a s u b s t i t u t e to an automobile would p o s i t i v e l y a f f e c t automobile ownership. T h i s i m p l i e s that i n f r e q u e n t or u n r e l i a b l e p u b l i c t r a n s p o r t a t i o n c h o i c e s would i n c r e a s e the l e v e l of ownership. F i n a l l y , a consumer's choice of an automobile depends on the set of automobiles a v a i l a b l e i n the market p l a c e . There are three b a s i c types of needs -- t r a v e l comfort and convenience, and a f f e c t i v e needs -- which consumer's expect to 59 r e c e i v e from an automobile. F i r s t , consumers expect automobiles to p r o v i d e safe and r e l i a b l e t r a n s p o r t a t i o n . Second, consumer's demand a c e r t a i n l e v e l of comfort, convenience and f l e x i b i l i t y from car usage. Given the f a c t that an automobile i s the second l a r g e s t consumer a s s e t , i t a l s o serves to s a t i s f y a f f e c t i v e or symbolic needs f o r i t s owners. The review of s t u d i e s i n s e c t i o n 2.4, suggested that income, education, race, household composition, l o c a t i o n of house and employment s t a t u s a f f e c t e d ownership d e c i s i o n s . A summary of marketing r e l a t e d s t u d i e s suggest that a t t i t u d e s and p s y c h o l o g i c a l d i s p o s i t i o n s may a f f e c t the c h o i c e of an automobile. However, the e f f e c t s of p s y c h o l o g i c a l f a c t o r s must be c o n s i d e r e d t e n t a t i v e at t h i s time. I n t e r r e l a t ions among b a s i c f a c t o r s . The ba s i c f a c t o r s d i s c u s s e d above provide a l i s t of f i r s t order e f f e c t s on ownership. However, i n t e r r e l a t i o n s among these f a c t o r s may a l s o generate second-order e f f e c t s . For example, t r a n s p o r t a t i o n planners have argued that p r e f e r e n c e s f o r p u b l i c t r a n s p o r t a t i o n depends upon consumers' p e r c e p t i o n of t h e i r own needs f o r convenience, performance, p e r s o n a l environment, r i d e q u a l i t y , s a f e t y , f l e x i b i l i t y , crowding, comfort and p r i c e ( W a l l i n and Wright, 1974; Recker and Golob, 1976). Thus, the model proposed must account f o r i n t e r r e l a t i o n s among market f a c t o r s and consumer demands. S i m i l a r l y , c o - v a r i a t i o n between consumer c h a r a c t e r i s t i c s and market f a c t o r s p rovide a b a s i s f o r understanding consumers' s o c i a l or economic c o n s t r a i n t s . 60 Fu r t h e r , consumer demands f o r comfort and s t a t u s and consumer c h a r a c t e r i s t i c s provides i n d i c a t i o n s about consumers' s a t i s f a c t i o n with t h e i r c u r r e n t l y owned c a r s . Of course s a t i s f a c t i o n , i n tu r n , a f f e c t s the f u t u r e c h o i c e of an automobile. The model uses some of the fundamental r e l a t i o n s h i p s from past r e s e a r c h f i n d i n g s and pr o v i d e s a broader framework f o r an i n v e s t i g a t i o n of the car ownership phenomena. Past r e s e a r c h , however, has not focused on the i n t e r r e l a t i o n s among the b a s i c f a c t o r s . For example, i t was argued i n s e c t i o n 2.4 that income has a p o s i t i v e impact on the number and s i z e of c a r s owned, while the p r i c e of g a s o l i n e has a negative impact. The c r o s s - e l a s t i c i t y e f f e c t s , however, among income and g a s o l i n e p r i c e s and automobile choice i s not w e l l understood. S i m i l a r l y , i t can be argued that comfort seekers with a dominant p e r s o n a l i t y , may own e i t h e r a l a r g e car or a small car with a powerful engine. Although i l l u s t r a t i n g a l l p o s s i b l e i n t e r r e l a t i o n s among the b a s i c f a c t o r s and t h e i r impact on ownership i s beyond the purpose of t h i s d i s c u s s i o n , an example of the i n t e r r e l a t i o n s between consumer c h a r a c t e r i s t i c s and automobile f l e e t c h a r a c t e r i s t i c s i s presented i n Table 2.4. For example, i t i s argued that the passenger c a r r y i n g c a p a c i t y and the number of household members provide an i n d i c a t i o n r e g a r d i n g household t r a v e l f l e x i b i l i t y . However, weekly t r a v e l expenditure ( g a s o l i n e p r i c e , t o t a l t r a v e l and v e h i c l e MPG) and income 61 Table 2.4 Automobile and consumer c h a r a c t e r i s t i c s i n t e r - r e l a t i o n s (Expanded from Manski and Sherman, 1980) Car c h a r a c t e r i s -t i c s Phys i c a l measure B a s i s f o r r e l a t i o n s h i p Consumer c h a r a c t e r i s t i c s measures Passenger c a r r y i n g capac i ty Seat ing space, Ride q u a l i ty Expected number of passengers c a r r i e d . Expected r i d e hours i n s i d e car Number of c h i l d r e n , household s i z e , work d i s t a n c e Load c a r r y -ing c a p a c i t y Volume or number of bag space, payload l i m i t s for van or t rue k Intended mix of uses, e.g. moving t h i n g s vs. urban commuting Percentage times car used at work, p r o f e s -s i o n Per formance A c c e l e r a t -ion , brak-ing and maneuvering q u a l i t y . Mechan i c a l r e l i a b i l i t y HP per l b Turn ing c i r c l e , Fa i l u r e p r o b a b i 1 i -t i e s Mix of c i t y and highway d r i v i n g , V u l n e r a b i l i t y of consumer to car u n a v a i l a b i l i t y f o r f o r short time Percentage d r i v i n g on c i t y roads Cost Purchase pr i c e , F u e l c o s t , Deprec i a t -ion r a t e Consumer c o n s t r a i n t to c o s t f a c t o r s e.g. Wealth, income s t a t u s Income, v e h i c l e m i l e s t r a v e l l e d C l a s s and , s t y l e , body -shape, "image" Number of s i m i l a r models in market, number of un ique f e a t u r e s to car P r e f e r e n c e s , a b i l i -ty to purchase "image" Pr e f e r e n c e to n o v e l t y or a n t i q u e s . Sensation seeking. Per c a p i t a earning or wealth 62 probably determine the economic c o n s t r a i n t s i n car usage. Thus, the i n t e g r a t e d model suggests that consumers' c h o i c e s of automobiles depend upon market f a c t o r s , consumer e x p e c t a t i o n s and consumer c h a r a c t e r i s t i c s . F u r t h e r , i n t e r r e l a t i o n s among these b a s i c f a c t o r s generates a p e r c e p t i o n about a l t e r n a t i v e s , c o n s t r a i n t s and s a t i s f a c t i o n s . Although the past r e s e a r c h reviewed i n the pre v i o u s s e c t i o n s p r o v i d e s some in f o r m a t i o n about each of the ba s i c f a c t o r s , l i t t l e i s known about the i n t e r r e l a t i o n s h i p s among them and the impacts of these r e l a t i o n s h i p s on ownership. 2.7 P o s s i b l e Research Areas There are two major focuses f o r f u r t h e r r e s e a r c h . F i r s t , there are s e v e r a l a n a l y t i c a l i s s u e s a s s o c i a t e d with automobile ownership m o d e l l i n g . Second, there i s a l s o a set of i s s u e s a r i s i n g from measurement and the use of a p p r o p r i a t e p r e d i c t o r s . In t h i s s e c t i o n , each of these concerns w i l l be reviewed. A n a l y t i c a l i s s u e s . There i s a t h e o r e t i c a l d i s t i n c t i o n between d i s c r i m i n a n t and l o g i t - t y p e d i s c r e t e c h o i c e models. For example, the e r r o r s t r u c t u r e a s s o c i a t e d with d i s c r i m i n a n t a n a l y s i s i s a normal d i s t r i b u t i o n and while f o r the l o g i t model i t i s log-normal. T h i s means that there may be a s u b s t a n t i a l p o s s i b i l i t y of reaching i n c o r r e c t c o n c l u s i o n s i f an i n a p p r o p r i a t e model i s a p p l i e d to observed data. A Monte-Carlo s i m u l a t i o n as w e l l as t h e o r e t i c a l a n a l y s i s may pro v i d e f u r t h e r 63 i n s i g h t s about the c h o ice of a model for e m p i r i c a l r e s e a r c h . The i n t e g r a t e d model proposed in the p r e v i o u s s e c t i o n suggests that ownership choice r e s u l t s from s e v e r a l f a c t o r s . However, past e m p i r i c a l work has focused on a l i m i t e d set of these f a c t o r s . More s p e c i f i c a l l y , i t was argued that e i t h e r demand f o r comfort or p e r c e p t i o n of comfort among the a l t e r n a t i v e s a f f e c t s ownership. Thus, there i s e i t h e r a d i r e c t impact from the b a s i c f a c t o r s and/or a second order impact on ownership. Future r e s e a r c h may p rovide a d d i t i o n a l i n s i g h t s about the nature of t h i s r e l a t i o n s h i p . The Lave and T r a i n (1979) study and the model in s e c t i o n 2.6 suggested that car d e s c r i p t o r v a r i a b l e s and consumer c h a r a c t e r i s t i c s give i n d i c a t i o n s about the l e v e l of s o c i a l or economic c o n s t r a i n t s . The c o n s t r a i n t s , i n t u r n , a f f e c t ownership c h o i c e s . Future r e s e a r c h may look f o r the e f f e c t s of p e r c e p t i o n and s a t i s f a c t i o n on ownership and automobile c h o i c e s . Measurement of ownership. The past research has mainly c o n c e n t r a t e d on ownership in terms of the number of c a r s owned, or in terms of a few automobile c h a r a c t e r i s t i c s . However, measurement and e s t i m a t i o n of v a r i o u s car c h a r a c t e r i s t i c s d i s c u s s e d i n s e c t i o n 2.3 and consumer c h a r a c t e r i s t i c s could provide a f u r t h e r understanding of the consumer c h o i c e process. Although past r e s e a r c h has not i n d i c a t e d the e f f e c t s of a t t i t u d e on the number of automobiles h e l d by households, i t might be expected that a t t i t u d e s a f f e c t s p e c i f i c car c h a r a c t e r i s t i c s as w e l l as the number of c a r s h e l d . Future 64 re s e a r c h may provide a t e s t of t h i s h y p o t h e s i s . S i m i l a r l y , r e s e a r c h may a l s o provide a r e p l i c a t i o n of the e f f e c t s of income, f a m i l y composition, employment, house l o c a t i o n , and education on ownership c h o i c e s . On the other hand, impact of fa m i l y l i f e - c y c l e , g a s o l i n e p r i c e , t r a v e l needs and housing choice on consumer automobile ownership have had only l i m i t e d success i n the research reviewed here. Future r e s e a r c h may be s u c c e s s f u l i f these r e l a t i o n s h i p s are r e p o s t u l a t e d , measured c o r r e c t l y and then t e s t e d with an a p p r o p r i a t e a n a l y t i c a l model. 65 CHAPTER III Data sources, Hypothesis and A n a l y t i c a l Methods. 3.1 Introduct ion 3.2 ECCP Sample to P o p u l a t i o n Estimate Comparisons 3.3 Mo d e l l i n g Number of Cars Owned 3.4 Mo d e l l i n g Car S i z e Preference 3.5 Methods of A n a l y s i s 66 3.1 I n t r o d u c t i o n One data set p a r t i c u l a r l y s u i t e d to studying consumer automobile ownership c h o i c e s i s the Energy Consumption and C o n s e r v a t i o n P a t t e r n (ECCP) sample d e s c r i b e d i n McDougall e t . a l . ( l 9 8 0 ) . T h i s data base c o n t a i n s e x t e n s i v e i n f o r m a t i o n about 2366 Canadian households, i n c l u d i n g consumer a t t i t u d e s towards energy usage. The purpose of t h i s chapter i s twofold. F i r s t , the ECCP sample estimates of average car ownership (number per household) and automobile performance c h a r a c t e r i s t i c s (miles per g a l l o n , MPG) are compared to the estimates of these v a r i a b l e s f o r the Canadian p o p u l a t i o n provided by S t a t i s t i c s Canada. Second, s p e c i f i c models to be t e s t e d in the next chapter are a l s o d e s c r i b e d . The ECCP sample to S t a t i s t i c s Canada data comparisons presented i n s e c t i o n 3.2 suggests that the ECCP sample o v e r s t a t e s the average MPG. F u r t h e r , f a m i l i e s without c a r s appear to be under-represented in the ECCP sample. F i n a l l y , the evidence f o r the number of c a r s owned per ho.usehold i s i n c o n c l u s i v e to decide r e p r e s e n t a t i v e n e s s of the ECCP sample. The a n a l y s i s component of t h i s t h e s i s was done i n three stages. The f i r s t stage ( s e c t i o n 3.3) was to attempt to r e p l i c a t e p r i o r models of number of c a r s owned. T h i s a n a l y s i s i s termed the c o n f i r m a t i o n model. More s p e c i f i c a l l y , the c o n f i r m a t i o n model was proposed to t e s t the e f f e c t s of income, 67 household composition, wife's employment s t a t u s , g a s o l i n e p r i c e , a v a i l a b i l i t y of p u b l i c t r a n s p o r t a t i o n , and l i f e - c y c l e stage on the number of c a r s owned. In the second stage, the e x p l o r a t o r y model extended the c o n f i r m a t i o n model to t e s t e f f e c t s of a t t i t u d e and t r a v e l need v a r i a b l e s . In the t h i r d stage ( s e c t i o n 3.4), a model was proposed to e x p l a i n the automobile s i z e (average car weight w i t h i n the household's f l e e t ) . I t was expected that income and household s i z e would have a p o s i t i v e impact on t h i s v a r i a b l e and that the p o p u l a t i o n s i z e of the c i t y i n which a household r e s i d e s , the consumer's age and h i s or her energy consciousness would have a negative impact on the average weight. In s e c t i o n 3.5, d e t a i l s of s t a t i s t i c a l techniques are summarized. The model fo r the number of c a r s owned was based on l o g i t and d i s c r i m i n a n t a n a l y s i s . The model fo r the average f l e e t weight was based on r e g r e s s i o n a n a l y s i s . 3.2 The ECCP Sample to the P o p u l a t i o n Estimate Comparisons. The t a r g e t p o p u l a t i o n f o r t h i s study was Canadian households that own one or more automobiles. The purpose of t h i s s e c t i o n i s to a s c e r t a i n the r e p r e s e n t a t i v e n e s s of the ECCP sample to the t a r g e t p o p u l a t i o n i n terms of the mean number of c a r s owned, new car purchasing p a t t e r n s and the average MPG achieved. For comparative purposes, the t a r g e t p o p u l a t i o n estimates were obtained from v a r i o u s S t a t i s t i c s Canada r e p o r t s . 68 In terms of general comparison, however, i t was al r e a d y r e p o r t e d by R i t c h i e e t . a l . (1981) that f a m i l i e s in the ECCP "...sample had higher incomes, a higher p r o p o r t i o n of two-adult f a m i l i e s , and a lower p r o p o r t i o n of second spouse working". Table 3.1 r e p o r t s the comparison between the ECCP sample and the p o p u l a t i o n estimates of the average number of c a r s per household. The p o p u l a t i o n estimates were based on (1) passenger car r e g i s t r a t i o n , (2) a household equipment survey of about 20,000 households, and (3) a s p e c i a l passenger car f u e l consumption study based on a monthly r o l l i n g sample of 1000 c a r s ; a l l reported by S t a t i s t i c s Canada. The ECCP estimate i n f e r r e d that the p o p u l a t i o n owned 1.61 c a r s per household while S t a t i s t i c s Canada samples r e p o r t e d an average ownership between 1.23 and 1.69 c a r s . Although d i f f e r e n c e s between the samples were s t a t i s t i c a l l y s i g n i f i c a n t at p^0«0l, the d i r e c t i o n of d i f f e r e n c e s depended upon the c h o i c e of p o p u l a t i o n estimates compared. More s p e c i f i c a l l y , the passenger car r e g i s t r a t i o n based estimate and the ECCP estimate d i d not show any major d i f f e r e n c e s or p a r t i c u l a r p a t t e r n a c r o s s the p r o v i n c e s . The passenger car f u e l consumption study and the household equipment survey, however, i n d i c a t e d a lower number of c a r s owned than the ECCP sample. These comparisons i n terms of c a r s per household, however, p r o v i d e d i n c o n c l u s i v e evidence f o r judging the ECCP sample's r e p r e s e n t a t i v e n e s s . In Table 3.2, a comparison between estimated p r o v i n c i a l new car s a l e s and the ECCP sample new c a r s , suggested that the 6 9 T a b l e 3.1 Average Car Ownership per Car Owning Household ECCP sample S t a t i s t i c s Canada t - s t a t . P r o v i n c e e s t imate p o p u l a t i o n e s t imate t o compare ECCP and A B C e s t i m a t e A Newfoundland 1 . 57 1 .31 1 . 27 - 1 . 08 1.11 P r i n c e Edward I s l a n d 1 .83 1 . 66 1 .61 1 . 56 0.93 Nova S c o t i a 1 .49 1 . 62 1 .28 1 . 30 -3 . 58 New Brunswick 1 . 53 1 .61 1 . 32 1 .37 -1.46 Quebec 1.71 1 . 34 1 .25 1 . 02 +19.81 O n t a r i o 1 .65 1 . 64 1 .39 1 .30 0.48 Man i toba 1 . 67 1 . 78 1 . 33 1.21 -2.14 Saskatchewan 1 .83 1 . 62 1 . 39 1 .32 + 3 . 24 A l b e r t a 1 .80 1 .92 1 . 48 1 . 53 -2.36 B r i t i s h Columbia 1 .83 1 .77 1 .46 1.15 1 .29 Sample 1.61 1 . 69 1 . 36 1 .23 -6.39 E s t i m a t e A was o b t a i n e d from the r a t i o of the t o t a l a u t o m o b i l e s r e g i s t e r e d ( S t a t i s t i c s Canada p u b l i c a t i o n , Road Motor V e h i c l e s R e g i s t r a t i o n s , C a t a l o g 53-219) and the t o t a l number of h o u s e h o l d s owning one or more a u t o m o b i l e s ( S t a t i s t i c s Canada p u b l i c a t i o n , Household F a c i l i t i e s and Equipment, May 1978, C a t a l o g u e 64-202). E s t i m a t e B was o b t a i n e d by m u l t i p l y i n g the number of h o u s e h o l d s owning one, two, and t h r e e or more c a r s w i t h numbers one, two, and t h r e e r e s p e c t i v e l y , then d i v i d i n g by number of f a m i l i e s owning one or more c a r s . E s t i m a t e C was o b t a i n e d by d i v i d i n g the number of P e r s o n a l Use Passenger c a r s o b t a i n e d from Passenger F u e l Consumption Survey, 1979, by the number of f a m i l i e s owning one or more c a r s . io T a b l e 3.2 Sample New Car Purchases and P o p u l a t i o n New Car S a l e s Number of new E x p e c t e d P r o v i n c e c a r s p u r c h a s e d number of new x 2 i n ECCP sample c a r s p u r c h a s e d 1 (% of t o t a l ) (market share) Newfoundland 3 (0.69) 7.3 ( 1 . 68) 2 . 53 P r i n c e Edward I s l a n d 6 (1 .38) 2.1 (0. 48) 2 .70 Nova S c o t i a 21 (4.83) 15.5 (3. 56) 1 .40 New Brunswick 13 (2.99) 11.5 (2. 64) 0 .20 Quebec 102(23.45) 118.5(27. 24) 1 .51 O n t a r i o 184(42.30) 172.2(39. 59) 0 .81 M a n i t o b a 23 (5.29) 17.0 ( 3 . 91 ) 2 . 1 2 Saskatchewan 19 (4.37) 14.0 (3. 22) 1 .34 A l b e r t a 31 (7.13) 37.1 (8. 53) 1 .00 B r i t i s h Columbia 33 (7.59) 39.9 (9. 17) 1 .09 T o t a l c a r s p u r c h a s e d 435(100) 435(100) 1 4 . 70 T o t a l c a r s s o l d i n Canada 988,890 1 E x p e c t e d number of c a r s a l e s was o b t a i n e d by m u l t i p l y i n g sample s i z e ( 4 3 5 ) t o market share f o r new c a r s . P r o v i n c i a l market share o b t a i n e d from S t a t i s t i c s Canada p u b l i c a t i o n , New Car S a l e s , 1978, C a t a l o g u e 53-169. 71 d i f f e r e n c e s between the two p o p u l a t i o n estimates were minimal. A goodness of f i t t e s t used to measure the d i f f e r e n c e s between the observed and the expected p r o p o r t i o n s of c a r s purchased i n d i c a t e d x 2 of 14.7 with nine degrees of freedom ( r e j e c t s the n u l l h ypothesis at p^O-10). In other words the new car purchases i n the ECCP sample c l o s e l y matched new car s a l e s i n the p o p u l a t i o n . The ECCP sample provided s e l f - r e p o r t s of MPG f o r each household c a r . Thus, a comparison between the S t a t i s t i c s Canada f u e l consumption study and the ECCP sample was examined. Although bqth samples were drawn from the same p o p u l a t i o n , the ECCP estimate was obtained by consumer s e l f r e p o r t s , using the household as the sampling u n i t , and with data being c o l l e c t e d i n 1978. On the other hand, S t a t i s t i c s Canada employed d e t a i l e d d i a r i e s , using i n d i v i d u a l automobiles as a sampling u n i t and the data was c o l l e c t e d i n 1980. A q u a l i t a t i v e comparison between the two samples suggests that both methods produced convergent ( c o r r e l a t ion of 0.75) e s t i ma t e s of MPG (see Table 3.3). However, the d i a r y study r e p o r t e d lower percentage e r r o r s ( r a t i o of standard d e v i a t i o n to mean) than the s e l f report study. F u r t h e r , some estimates based on the s e l f r e p o r t s i n d i c a t e d a higher l e v e l of f u e l e f f i c i e n c y . These d i f f e r e n c e s d i d not appear a t t r i b u t a b l e to the time the sample was c o l l e c t e d (1978 and 1979) or to seasonal a v e r a g i n g . Since i t was expected that the d i a r y method would produce a more ac c u r a t e measure, i t was concluded that the ECCP sample over s t a t e d the MPG estimate. Table 3.3 F l e e t E f f i c i e n c y Comparisons ECCP sample F l e e t MPG(km per 1) e s t i m a t e * made i n 1980 MPG(km per 1) f o r 1978 and p r i o r manufactured c a r s P r o v i n c e estimate+ Q u a r t e r Annual average Oct-Dec Jan-Mar Apr-Jun J u l -Sep Newfoundland 17.65(6.25) 15.9(5 .6) 13 .7(4.9) 17.2(6.1) 18 . 4(6. 5) 16 .30(5. 78) P r i n c e Edward I s l a n d 19.56(6.93) 16.6(5 •9) 15 .7(5.6) 19.2(6.9) 20 . 8(7. 4) 18 .08(6. 45) Nova S c o t i a 19.76(6.99) 18.9(6 •7) 16 .5(5.8) 18.6(6.6) 20. 8(7. 4) 18 .70(6. 63) New Br u n s w i c k 19.02(6.73) 17.6(6 •2) 16 .1(5.7) 18.2(6.4) 20 . 9(7. 4) 18 .20(6. 43) Quebec 18.23(6.45) 15.9(5 .6) 14 .7(5.2) 17.8(6.3) 18 . 4(6. 5) 16 .70(5. 90) O n t a r i o 19.23(6.81) 17. 2 (6 • 1) 15 .1(5.4) 18.4(6.5) 18 . 9(6. 7) 17 .40(6. 18) Manitoba 18.11(6.41) 15.2(5 .5) 14 .8(5.2) 18.9(6.7) 18 . 2(6. 5) 17 .08(5. 98) Saskatchewan 16.52(5.85) 14.9(5 • 3) 13 .9(4.9) 17.5(6.2) 18 . 0(6. 4) 16 .08(5. 70) A l b e r t a 18.24(6.46) 16.5(5 .8) 14 .9(5.3) 17.7(6.3) 18 . 8(6. 6) 16 .98(6. 0) B r i t i s h Columbia 18.40(6.51) 18.5(6 .5) 17 .4(6.2) 19.3(6.8) 21 . 1(7. 5) 19 .08(6. 75) T o t a l 18.65(6.60) 16.8(6 .0) 15 .3(5.4) 18.3(6.5) 19 . 1(6. 8) 17 .38(6. 18) + Obt a i n e d by a v e r a g i n g c i t y and highway MPG w i t h r e s p e c t i v e p r o p o r t i o n of d r i v i n g . o i * Ob t a i n e d from Passenger Car F u e l Consumption s u r v e y , v a r i o u s q u a r t e r s , S t a t i s t i c s Canada. 74 The f i n a l comparison was made between S t a t i s t i c s Canada's household equipment survey and the ECCP sample i n terms of p a t t e r n s of automobiles owned. I t appears from Table 3.4 that households without c a r s are l e s s frequent i n the ECCP sample than i n the S t a t i s t i c s Canada study. Since the t a r g e t p o p u l a t i o n of t h i s t h e s i s i s car owning households, t h i s d i f f e r e n c e i s not con s i d e r e d to be s e r i o u s . To summarize, p o p u l a t i o n estimates based on the ECCP sample of the MPG i s o v e r s t a t e d compared to estimates from S t a t i s t i c s Canada. F u r t h e r , the new car purchases in the ECCP sample c l o s e l y matched new car s a l e s i n the p o p u l a t i o n . The measures of the number of ca r s owned per household, however, provided i n c o n c l u s i v e evidence to judge r e p r e s e n t a t i v e n e s s of the ECCP sample. 3.3 M o d e l l i n g Number of Cars Owned In t h i s s e c t i o n two models f o r p r e d i c t i n g the number c a r s owned by a f a m i l y are d e s c r i b e d - the co n f i r m a t o r y model and the e x p l o r a t o r y model. The purpose of the co n f i r m a t o r y model was to estimate a model that was comparable to the models d e s c r i b e d i n the l i t e r a t u r e review. The e x p l o r a t o r y model extends the c o n f i r m a t o r y model to i n c l u d e a t t i t u d i n a l and the car r e l a t e d f a c t o r s . A d e s c r i p t i v e o u t l i n e of each of the models i s d i s c u s s e d below. Confirmatory Model: Cragg and Uhler (1970), Johnson (1978) and Table 3.4 Ownership P a t t e r n Comparisons - Number of Households P r o v i n c e No c a r owners One c a r owners Two c a r owners t h r e e or more c a r owners ECCP ECCP ECCP ECCP +Expected Sample Expected Sample Expected Sample Ex p e c t e d Sample Newfoundland 3.7 0 7 .9 9 2 1 3 0 . 3 2 P r i n c e Edward I s l a n d 4 . 4 1 14 . 0 10 5 1 10 1 . 5 4 Nova S c o t i a 29 . 1 12 59 . 5 53 16 8 40 3 . 6 4 New Brunswick 14 . 9 3 40 . 1 39 13 0 22 3 . 0 7 Quebec 157 . 0 85 350 . 0 368 79 0 127 16 . 0 22 O n t a r i o 173 . 0 44 466 . 0 368 195 0 342 37 . 0 84 Manitoba 28 . 0 4 69 . 0 59 23 0 46 4 . 0 15 Saskatchewan 16 . 0 1 50 . 0 32 18 0 43 6 . 0 14 A l b e r t a 26.0 5 88 . 0 62 43 0 82 13 . 0 21 B r i t i s h Columbia 50 . 0 13 129 . 0 87 66 0 129 16 . 0 32 Car owners o n l y 1622 • o 1120 588 0 844 126 . 0 205 + Expected f r e q u e n c i e s o b t a i n e d by m u l t i p l y i n g n o r m a l i z e d p r o p o r t i o n f o r each p r o v i n c e and sample s i z e s of t h a t p r o v i n c e . 77 T r a i n (1980) a l l found that the household income p o s i t i v e l y a f f e c t e d the number of c a r s owned. These s t u d i e s a l s o r e p o r t e d that the household s i z e had a p o s i t i v e a s s o c i a t i o n with the number of c a r s owned. F u r t h e r , Cragg and Uh l e r , and T r a i n r e p o r t e d that the l o c a t i o n of a consumer's re s i d e n c e r e l a t i v e to the c i t y - c e n t r e a f f e c t e d the number of ca r s owned. Since the ECCP data base d i d not provide an i n d i c a t i o n about house l o c a t i o n , a proxy v a r i a b l e , d i s t a n c e between the work and home, was t e s t e d . I t was expected that the g r e a t e r the work to home d i s t a n c e the higher the number of c a r s owned. Past research t e s t e d the l i n e a r e f f e c t of age of the household head on the number of c a r s owned, but f a i l e d to prov i d e c o n s i s t e n t r e s u l t s . Recent rese a r c h ( F r i t z s c h e , 1981), has suggested that the age of the household head may bear a no n - l i n e a r r e l a t i o n s h i p to car g a s o l i n e consumption and the number of ca r s owned. S p e c i f i c a l l y , t h i s t h e s i s proposed to t e s t the p r o p o s i t i o n that a middle aged household head with m u l t i p l e dependents would r e q u i r e a g r e a t e r number of automobiles than a r e t i r e d couple. S i m i l a r l y , a young couple would r e q u i r e greater f l e x i b i l i t y i n t r a v e l l i n g and thus would need a gr e a t e r number of automobiles than a r e t i r e d couple. Purchase d e c i s i o n s t u d i e s by Greenlees (1980) and Lave and T r a i n (1979) repo r t e d that g a s o l i n e p r i c e n e g a t i v e l y a f f e c t e d the s i z e of car purchased. T h i s suggested that the higher p r i c e d g a s o l i n e would l e a d to lower numbers of c a r s owned. The ECCP data base, however, d i d not have i n f o r m a t i o n on g a s o l i n e 78 p r i c e s . However, i t i s w e l l known that g a s o l i n e p r i c e s vary a c r o s s the regions w i t h i n the country, hence r e g i o n a l i n d i c a t o r v a r i a b l e s were used as a proxy to t e s t f o r the impact of g a s o l i n e p r i c e on the number of c a r s owned. T r a i n (1980) reported that the a v a i l a b i l i t y of p u b l i c t r a n s i t n e g a t i v e l y a f f e c t e d the number of c a r s owned. Since the ECCP data base d i d not i n d i c a t e a v a i l a b i l t y of p u b l i c t r a n s i t i t was assumed in t h i s study that consumers r e s i d i n g i n l a r g e r c i t i e s would be more l i k e l y to have access to p u b l i c t r a n s p o r t a t i o n . Thus, i t was expected that consumers l i v i n g in l a r g e r c i t i e s would own fewer c a r s per f a m i l y . To summarize, the c o n f i r m a t o r y model attempted to reproduce the e f f e c t of consumer c h a r a c t e r i s t i c s on the number of c a r s owned repor t e d i n other r e s e a r c h . These e f f e c t s of income, household composition and house l o c a t i o n are w e l l documented as r e p o r t e d i n chapter I I . E f f e c t s of f a m i l y l i f e - c y c l e , employment s t a t u s , r e g i o n a l v a r i a b l e s and a v a i l a b i l i t y of p u b l i c t r a n s p o r t a t i o n , however, were deduced from the past econometric r e s e a r c h . E x p l o r a t o r y Model: The c o n f i r m a t i o n model proposed above was t e s t e d by past r e s e a r c h e r s , however, the ECCP data base a l s o c o n t a i n e d energy c o n s e r v a t i o n a t t i t u d e s , and l i f e s t y l e d e s c r i p t o r s . The purpose of the e x p l o r a t o r y model i s to expand the c o n f i r m a t o r y model to i n c l u d e the e f f e c t of these f a c t o r s . Although importance of a t t i t u d e s in energy consumption was acknowledged by R i t c h i e e t . a l . , they d i d not i d e n t i t i f y any 79 s p e c i f i c p a t t e r n i n the ECCP data base. The f i r s t e x p l o r a t o r y model t e s t e d the impact of a t t i t u d i n a l f a c t o r s on the number of c a r s owned. It was expected that consumers conscious of t h e i r own g a s o l i n e consumption, would own fewer c a r s . S i m i l a r l y , consumers p o s i t i v e l y disposed towards energy c o n s e r v a t i o n and p u b l i c t r a n s p o r t a t i o n would be expected to own fewer automobiles. F i n a l l y , consumers a c t i v e l y p a r t i c i p a t i n g i n g a s o l i n e c o n s e r v a t i o n a c t i v i t i e s were a l s o l i k e l y to own fewer automobiles. Since an automobile p r o v i d e s t r a n s p o r t a t i o n s e r v i c e s , consumers with a greater demand on these s e r v i c e s were expected to own m u l t i p l e automobiles. More s p e c i f i c a l l y , consumers r e q u i r i n g to commute longer d i s t a n c e s to work and those needing a car at work, c o u l d be viewed as v u l n e r a b l e to car u n a v a i l a b i l i t y . Thus, to p r o t e c t themselves from automobile f a i l u r e s , consumers with longer weekly commuting d i s t a n c e s and consumers using a car at work would be expected to own a g r e a t e r number of c a r s per household than o t h e r s . To summarize, i t was expected that the e x p l o r a t o r y model would i n d i c a t e the r e l a t i v e importance of demographic, s i t u a t i o n a l , t r a v e l need, and a t t i t u d i n a l f a c t o r s on the number of c a r s owned. A " w h o l i s t i c " m o d e l l i n g approach such as t h i s has not yet been reported i n the l i t e r a t u r e . 80 3.4 M o d e l l i n g Car S i z e Preference The summary of past research on consumer p r e f e r e n c e s and car s i z e r e p o r t e d in the p r e v i o u s chapter suggested that s i z e p r e f e r e n c e s are s y s t e m a t i c . F u r t h e r , i n Chapter I, i t was argued that car s i z e or weight a l s o p r e d i c t s MPG. In other words, mod e l l i n g car s i z e can have important i m p l i c a t i o n s for i s s u e s concerning f u e l e f f i c i e n c y . The model proposed i n t h i s s e c t i o n p r e d i c t e d average automobile weight w i t h i n the fa m i l y u sing consumer t r a v e l needs (comfort, t r a n s p o r t a t i o n e t c . ) , consumer c h a r a c t e r i s t i c s , and a t t i t u d e s . Some of the expected r e s u l t s are summarized below. In terms of consumer t r a v e l needs, the consumer commuting longer d i s t a n c e s to and from work may f e e l the need fo r g r e a t e r comfort, and hence, a l a r g e r c a r . S i m i l a r l y , i f an automobile i s needed at the work p l a c e , to c a r r y payload, the consumer would be expected to own a l a r g e r c a r . For consumers r e s i d i n g in non-urban areas, r e g u l a r shopping t r i p s i n v o l v i n g bulky purchases f o r the household would be expected to r e q u i r e a l a r g e r c a r . In terms of the e f f e c t of demographic f a c t o r s on car s i z e , the consumer with higher income would be expected to own a l a r g e r automobile, s i n c e the income would f a c i l i t a t e the s t a t u s and d r i v i n g comfort a s s o c i a t e d with the ownership of l a r g e r and/or more expensive automobiles. S i m i l a r l y , i f a l a r g e r household, owns only one c a r , then i t would be expected to 81 p r e f e r a l a r g e r v e h i c l e . On the other hand, a l a r g e r household with both husband and wife employed r e q u i r e s g r e a t e r t r a n s p o r t a t i o n f l e x i b i l i t y , and i t would be achieved by owning m u l t i p l e c a r s , p o s s i b l y s m a l l e r ones. T h i s suggested that a l a r g e r f a m i l y would be l i k e l y to have at l e a s t one l a r g e c a r , while the s i z e of a d d i t i o n a l v e h i c l e s would depend upon the number of c a r s owned and the employment s t a t u s of the f a m i l y . F i n a l l y , i t was suspected that o l d e r consumers would be i n t e r e s t e d i n comfort and s a f e t y , and thus, would p r e f e r to own a l a r g e r c a r . In terms of the e f f e c t of a t t i t u d e s on car s i z e p r e f e r e n c e s , i t was a n t i c i p a t e d that consumers p a r t i c i p a t i n g i n g a s o l i n e c o n s e r v a t i o n behaviours would l i k e l y own smaller automobiles. The higher a consumer's awareness of g a s o l i n e expenses, the smaller i s he or she l i k e l y to own a c a r . F u r t h e r , the g r e a t e r a consumer's r e l u c t a n c e to pay the d i r e c t taxes a s s o c i a t e d with a l a r g e r car and accompanying higher g a s o l i n e consumption, the l a r g e r would be h i s or her c a r . To summarize, models proposed in t h i s s e c t i o n were designed to i n v e s t i g a t e s e v e r a l determinants of a consumer's c h o i c e of automobile s i z e . More s p e c i f i c a l l y , the c h o i c e of average car weight was modelled using demographic, a t t i t u d i n a l , and s i t u a t i o n a l v a r i a b l e s . I t was expected that income and household s i z e would have a p o s i t i v e e f f e c t on automobile s i z e (car weight), and c i t y ' s p o p u l a t i o n , the age of the household head. Energy consciousness would n e g a t i v e l y a f f e c t t h i s 82 dependent v a r i a b l e . 3.5 Methods of A n a l y s i s The purpose of t h i s s e c t i o n i s to summarize the a n a l y t i c a l methods used in reducing the data base s i z e , in model f o r m u l a t i o n , and i n t e s t i n g the r e l i a b i l i t y of the v a r i o u s models analyzed. Var i a b l e s c r e e n i n g : The ECCP data base c o n t a i n s over 500 v a r i a b l e s . Hence, judgement was r e q u i r e d to o b t a i n a manageable subset of v a r i a b l e s . House c h a r a c t e r i s t i c s and household a p p l i a n c e ownership v a r i a b l e s appeared i r r e l e v a n t f o r p r e d i c t i n g automobile c h o i c e . F u r t h e r , the ECCP data base c o n t a i n s husband and wife responses to energy consumption a t t i t u d e s and p a r t i c i p a t i o n items, however, only seven items from each of the sets ( a t t i t u d e and p a r t i c i p a t i o n ) appear r e l e v a n t to car usage (see Table 3.5). F u r t h e r , the husband's o p i n i o n s on these i s s u e s were found to be r e l a t e d to the wife's o p i n i o n s . Hence, only the husband's o p i n i o n s were used i n the subsequent a n a l y s i s . T h i s judgemental a n a l y s i s reduced the data set to 14 a t t i t u d e r e l a t e d v a r i a b l e s (Table 3.5), seven demographic and s i t u a t i o n a l f a c t o r s , and s i x automobile need r e l a t e d v a r i a b l e s (Table 3.6). A l l were used to p r e d i c t consumer car ownership c h o i c e s . A n a l y s i s Sequence: For a l l subsequent a n a l y s e s , the ECCP sample was d i v i d e d i n t o two subsamples: the t r i a l and the r e p l i c a t i o n 83 Table 3.5 A t t i t u d e and P a r t i c i p a t i o n Items from ECCP Data Base (sample s i z e s i n parentheses) Note: For the f i r s t seven items, response was on a s i x - p o i n t s c a l e with one denoting s t r o n g l y d i s a g r e e . For the l a s t seven items, each response was e i t h e r y e s ( l ) or no(2). Table 3.6 D e s c r i p t i o n of Independent V a r i a b l e s (except A t t i t u d e and p a r t i c i p a t i o n items) 8 4 V a r i a b l e A Measure of S t d . Dev. Sample C e n t r a l S i z e Tendency Demographic F a c t o r s Income ( i n '000). F a m i l y s i z e . Husband's age ( y e a r s ) . W i f e ' s age ( i n y e a r s ) . Model l i f e - c y c l e group. Pecept of female heads of f a m i l y employed o u t s i d e . Modal Husband's e d u c a t i o n . Modal Husband's o c c u p a t i o n 1 7, 3, 48, 46, 93 44 4 1 09 9. 1 15 1 4 1 1 36 24 80 Young c o u p l e -NA-youngest c h i l d l e s s than 4 y r s 37.80 -NA-secondary -NA-o f f i c e r e l a t e d - -NA-2357 2354 2149 2355 2345 2358 21 60 2358 S i t u a t i o n a l F a c t o r s Mean degree days(deg F ) . 8234.00 1486.25 2333 Average c i t y p o p u l a t i o n . 402.58 428.61 2358 P e r c e n t l i v i n g i n O n t a r i o . 37.20 -NA- 2366 P e r c e n t l i v i n g i n Quebec. 25.90 -NA- 2366 P e r c e n t l i v i n g i n s i n g l e f a m i l y d etached home. 76.60 -NA- 2323 T r a v e l Need R e l a t e d F a c t o r s Weekly commuting t o work ( m i l e s ) . P e r c e n t of d r i v i n g on c i t y r o a d s . Average f l e e t age ( i n y r s ) . P e r c e n t of f a m i l i e s u s i n g c a r ( s ) at work. P e r c e n t of f a m i l i e s owning a t l e a s t one a i r - c o n d i t i o n e d c a r . P e r c e n t of f a m i l i e s owning one or more r e c r e a t i o n a l i t e m s . 89.29 126.09 1626 52.50 -NA- 1866 4.18 2.99 2034 18.40 -NA- 1895 19.50 -NA- 2072 41.00 -NA- 2350 -NA- Not A p p l i c a b l e 8 5 D e s c r i p t o r s Male sample mean sc o r e Female sample mean sc o r e 1 . I would be w i l l i n g t o pay d o u b l e the c u r r e n t p r i c e f o r g a s o l i n e i f I thought a l l Canadians would use l e s s 1 . 66 (1891) 1 .86 (2304) 2. I l i k e d r i v i n g new c a r s b e t t e r than o l d ones. 4.18 (1869) 3.71 (2231 ) 3. People s h o u l d not be a l l o w e d t o d r i v e t h e i r c a r s t o work u n l e s s they have a t at l e a s t one passenger. 2.66 ( 1 894) 2.84 (2319) 4 . A u t o m o b i l e m a n u f a c t u r e r s s h o u l d not be a l l o w e d t o produce c a r s t h a t get l e s s than 25 MPG. 4.85 ( 1891) 4.90 (2308) 5. People who own l a r g e c a r s s h o u l d have to pay $300 a year f o r l i c e n s e p l a t e s . 2.61 (1895) 2.54 (2310) 6. Everyone ( i n c l u d i n g me) who l i v e s near bus r o u t e s , s h o u l d take the bus i f i t g e t s them c l o s e t o where they want t o go. 4.12 ( 1 886) 4.55 (2303) 7. A subcompact c a r c o u l d meet my t r a n s p o r t a t i o n needs. 4.24 (1883) 4.74 (2274) 8. The p r e s e n t c a r I own i s s m a l l e r than the l a s t c a r I owned. 1 .60 (1770) 1 .62 (1778) 9. In the l a s t t w e l v e months, I have had my c a r tuned up a t l e a s t t w i c e . 1 .36 (1789) 1 .38 (1793) 1 0 . I u s u a l l y take the bus t o work. 1.91 (1390) 1 .82 (1054) 1 1 . I keep a r e c o r d of the amount 1^  spend on g a s o l i n e f o r my c a r . 1 . 58 (1812) 1 .62 (1788) 1 2 . I keep a r e c o r d of the number of m i l e s my c a r i s d r i v e n each y e a r . 1 .44 (1809) 1 .83 (1798) 1 3 . I u s u a l l y take the bus t o go s h o p p i n g . 1 . 94 (1540) 1 .82 (1795) 1 4 . In the l a s t y e a r , I went on a v a c a t i o n t r i p , by c a r , over 500 m i l e s from home 1 .61 (1881 ) 1 .62 (2269) 86 sub-samples. The t r i a l sample was used to estimate s e v e r a l models that p r e d i c t the p r o b a b i l i t i e s of a household owning one, two or three automobiles, and the average car weight. The r e p l i c a t i o n sample was used to reproduce the r e s u l t s of the "best" model from the t r i a l sample. The f i n a l model chosen f o r p r e d i c t i n g the p r o b a b i l i t i e s of the number of c a r s owned i s based on a l o g i t and a d i s c r i m i n a n t a n a l y s i s . In each a n a l y s i s , the subset of v a r i a b l e s (demographic, s i t u a t i o n a l , e t c . ) was t e s t e d f o r t h e i r r e l a t i v e a b i l i t y to p r e d i c t the number of c a r s owned. V a r i a b l e s with l e s s than one t - s t a t i s t i c i n a b s o l u t e value (p^0«25), were co n s i d e r e d u n s u i t a b l e f o r the f i n a l model. A c o l l e c t i o n of s u i t a b l e v a r i a b l e s from a l l subsets, were then used to estimate the f i n a l model. T h i s l a s t s tep was repeated u n t i l a l l v a r i a b l e s appeared s i g n i f i c a n t at p<0-25. Although the procedure i s not optimal i n terms of the percentage of v a r i a t i o n e x p l a i n e d ( R 2 ) , i t provided c o n s i s t e n t estimates and r e q u i r e d no s p e c i a l purpose software. The model developed f o r average car weight was f i t t e d by procedures i n v o l v i n g forward stepwise r e g r e s s i o n (Draper and Smith, 1966) and an a l l - s u b s e t r e g r e s s i o n a n a l y s i s (Mallows, 1973). The v a r i a b l e s with s i m i l a r signs with the c o e f f i c i e n t s i g n i f i c a n t at l e a s t at the P<0«10 l e v e l , were used i n the f i n a l model. I t was expected that low i n t e r - c o r r e l a t i o n s e x i s t between the independent v a r i a b l e s and that t h i s v a r i a b l e s e l e c t i o n procedure would l i k e l y produce c o n s i s t e n t r e s u l t s 87 (Hocking, 1976). The estimated parameters f o r v a r i o u s models along with t h e i r i m p l i c a t i o n s are d i s c u s s e d and summarized in the next chapter. 88 CHAPTER IV R e s u l t s and C o n c l u s i o n . 4 . 1 Introduct ion 4.2 Models R e l a t i n g to Number of Cars Owned 4.3 Models P r e d i c t i n g Average Car Weight 4.4 Comprehensive model of Car F l e e t Choice 4.5 P o l i c y I m p l i c a t i o n s and Future Research 89 4.1 I n t r o d u c t ion The purpose of t h i s chapter i s to summarize and comment on the a n a l y s i s of the v a r i o u s models c o n s t r u c t e d f o r p r e d i c t i n g (1) the number of cars owned by Canadian f a m i l i e s , and (2) the average car weight. In s e c t i o n 4.2, the models r e l a t i n g to the number of c a r s owned i n d i c a t e s that income, f a m i l y s i z e , and employment s t a t u s a l l p o s i t i v e l y a f f e c t e d the number of c a r s . The e f f e c t of the fami l y l i f e - c y c l e stage i s n o n - l i n e a r and complex. F u r t h e r , the p r o x i e s f o r g a s o l i n e p r i c e s , and a resid e n c y i n l a r g e urban c e n t r e s have the expected negative e f f e c t on the number of c a r s owned. In s e c t i o n 4.3, models r e l a t i n g to the average weight of car s owned by the Canadian household suggest that a t t i t u d e s towards and p a r t i c i p a t i o n i n g a s o l i n e c o n s e r v a t i o n a c t i v i t i e s were strong p r e d i c t o r s of t h i s v a r i a b l e . Moreover, ownership of t r a n s p o r t a t i o n r e l a t e d r e c r e a t i o n a l items r e q u i r e s consumers to own bigger (heavier) c a r s . I t i s a l s o noted that l a r g e r f a m i l i e s tend to own l a r g e r c a r s . An attempt i s made in s e c t i o n 4.4 to i n t e g r a t e the model of the number of c a r s owned and the model of the average car weight to p r o v i d e a more comprehensive d e s c r i p t i o n of the determinants of household f l e e t composition. I t i s concluded' that f l e e t c hoice i s contingent on demographic f a c t o r s , s i t u a t i o n a l f a c t o r s , t r a v e l needs and a t t i t u d i n a l f a c t o r s . F u r t h e r , i t i s argued that demographic and s i t u a t i o n a l f a c t o r s s t r o n g l y a f f e c t 90 the number of c a r s owned, while t r a v e l needs and a t t i t u d i n a l f a c t o r s do a b e t t e r job i n p r e d i c t i n g f l e e t weight. In s e c t i o n 4.5, p o l i c y i m p l i c a t i o n s f o r energy c o n s e r v a t i o n and t o p i c s f o r f u r t h e r modelling are d i s c u s s e d and summarized. It i s suggested that p o l i c i e s based on g a s o l i n e p r i c e s or p u b l i c p o l i c i e s based on the a b i l i t y to pay may have l i m i t e d success i n conserving g a s o l i n e . However, p o l i c i e s that encourage the use of e f f e c t i v e and e f f i c i e n t p u b l i c t r a n s p o r t a t i o n systems are l i k e l y to have m u l t i p l e b e n e f i t s i n c l u d i n g g a s o l i n e c o n s e r v a t i o n . I t i s a l s o suggested that f u t u r e m o d e l l i n g work co u l d examine car usage behaviour and the dynamics of car f l e e t change. F u r t h e r , a f u t u r e attempt to model g a s o l i n e consumption should be made " w h o l i s t i c " , using c a u s a l s t r u c t u r e which i s suggested i n t h i s study. 4.2 Models R e l a t ing to the Number of Cars Owned In t h i s s e c t i o n , models c o n s t r u c t e d to p r e d i c t the p r o b a b i l i t i e s a s s o c i a t e d with consumer ownership of one, two, three or more v e h i c l e s w i l l be summarized. The f i r s t model d i s c u s s e d u t i l i z e s only demographic and s i t u a t i o n a l v a r i a b l e s , while the second model uses a t t i t u d e , energy c o n s e r v a t i o n r e l a t e d behaviours, and t r a v e l need r e l a t e d v a r i a b l e s . Both models are " s t a t i s t i c a l l y s i g n i f i c a n t " and t h e i r d e s c r i p t i o n of consumer behaviour appears c o n c e p t u a l l y c o n s i s t e n t . Although only the r e s u l t s of the c o n f i r m a t o r y and the e x p l o r a t o r y l o g i t 91 a n a l y s i s are reported here, d i s c r i m i n a n t a n a l y s i s provided s i m i l a r r e s u l t s . Confirmatory model The r e s u l t s of maximum l i k e l i h o o d e s t i m a t i o n f o r the l o g i t model are presented in Table 4.1. The model t e s t e d a l l demographic v a r i a b l e s , and showed that income, household s i z e and wife's employment s t a t u s had a p o s i t i v e impact on the number of c a r s owned. Moreover, e i g h t out of ten parameters r e l a t e d to l i f e - c y c l e i n d i c a t o r s were s t a t i s t i c a l l y s i g n i f i c a n t . Among the other demographic v a r i a b l e s , the parameters a s s o c i a t e d with age of household head, middle age f a m i l y with youngest c h i l d l e s s than four years o l d , and head of household r e t i r e d d i d not show s t a t i s t i c a l l y s i g n i f i c a n t r e s u l t s . The model that t e s t e d r e g i o n a l v a r i a b l e s and s i t u a t i o n a l f a c t o r s , i n d i c a t e d that households in e a s t e r n Canada owned fewer v e h i c l e s than those in the western r e g i o n . In a d d i t i o n , households owned fewer c a r s per household i f they r e s i d e d i n " l a r g e " c i t i e s . The r e g i o n a l d i f f e r e n c e s w i t h i n the western re g i o n and the "severness of c l i m a t e " were not s t a t i s t i c a l l y s i g n i f i c a n t . The f i n a l model chosen f o r comparative purposes was d e f i n e d with 15 v a r i a b l e s . A s a t i s f a c t o r y t e s t of the s t a t i s t i c a l s i g n i f i c a n c e of the e n t i r e model i s to compare the r a t i o of the value of the l o g - l i k e l i h o o d f u n c t i o n with a l l parameters estimated to the value of the l o g - l i k e l i h o o d f u n c t i o n under the c o n s t r a i n t that a l l c o e f f i c i e n t s except the i n t e r c e p t terms are Table 4.1 MLE for P r e d i c t i n g P r o b a b i l i t i e s of Number of Cars Owned Demographic and S i t u a t i o n a l Variables ( t - s t a t i s t i c s i n parentheses) T r i a l sample consumer choice Two(0) Two(0) Two (0)- Two(0) Two(0) Two(0) Var i a b l e s vs. vs. vs. vs. vs vs. One(1) Three(1) One(1) Three(1) One(1) Three(1) car(s) cars car(s) cars car(s) cars Income ( i n '000) -0 056 -0 0119 -0 0715 0. 0395 -0. 063 0 .0143 (-5 89) (-7.17) (2 .65) (-9 20) (1.39) Household s i z e -0 191 0 611 -0 249 0. 442 -0 219 0 . 51*8 (-2 49) (5 17) (-3 22) (3. 99) (-4 04 ) (6 . 55) If wife i s employed -0 191 0 496 -0.0232 0.0452 -0 116 0 . 230 (0=No; l=yes) (-1 24) • (1 89) (-1 04) (1 .30) Young couple - No -0 492 1 237 -1 550 -7 . 798 -0 950 0.0644 c h i l d r e n (-1 44 ) (1 60) (-4 03) (-3 76) Young family with -0 520 -0 317 -0 581 -0 . 776 -0 543 -0 . 606 youngest c h i l d below (-2 04) (-2 21) (-1. 43) (-2 98) (-1 .50) 4 years Young family with -0 561 0 421 -0 893 -0. 453 -0 709 -0. D699 youngest c h i l d above (-1 47) (-2 35) (-2 65) 4 years Middle age family with -0 766 1 171 -0 656 0. 705 -0 729 0. 394 no c h i l d r e n (-2 71) (1 67) (-2 11) (1. 11) (-3 51) (1 -91) Middle age family with -0 963 -0 410 -1 024 -0. 616 -0 979 -0 .538 youngest c h i l d (-3 30) (-3 47) (-1- 25) (-4 74) (-1 .42) between 4 to 8 yrs Middle age family with -0 768 1 089 -0 989 0 220 -0 864 0 .623 youngest c h i l d • (-2 73) (2 01) (-3 42) (-4 30) (1 .82) between 8 to 12 yrs Middle age family with -1 412 1 621 -1 981 1 304 -1 671 1 . 422 youngest c h i l d (-4 58) 1.7. 8f.) (-5 09) (2- 68) (-7 27 ) (3 .92) above 12 years Old age family with -0 674 1 747 -0 969 -0 j 11 -0 812 0 . 809 -some other family (-2 13) (2 70) (-3 14) (-3 69) (1 .78) member l i v i n g A t l a n t i c region 0 978 -0 521 0 728 -0. 358 0 830 -0 .434 (3 58) (2 68) (2 68) Quebec 1 451 -0 746 1 679 -1. 107 1 546 -0 .862 (7 06) (-1 93) (7 58) (-2. 49) (10 39) (-3 .03) Ontario 0 476 -0 220 0 613 -0 393 0 526 -0 .287 (2 73) (3 33) (-1 39) (4 17) (-1 . 50) C i t y s i z e 0.000596 0.000046 0.000788 -0.00065 0.000676 -0. 00031 (in '000) (3 28) (4 05) (-1 90) (5 11) (-1 . 36) Constant 1 802 -2 411 2 316 -3 441 2 054 -2 .979 (3 18) (-1 82) (4 82) (-1 11) (5 66) (-2 .08) Re p l i c a t i o n sample Consumer choice Total sample Consumer choice Pseudo Ra R* l i k e l i h o o d r a t i o t e s t cross v a l i d a t e d Ra 0.218 0 . 268 452.10 with 31 df . 0.282 0.251 0.309 503.8 with 31 df 0.247 0.226 0 . 282 923.09 with 31 df Sample d i s t r i b u t i o n s one car owners two car owners three car owners Total 553 431 104 1088 557 403 95 1055 1110 834 199 2143 A l l t - s t a t i s t i c s l e s s than one i n absolute value omitted. 93 equal to zero. T h i s r a t i o has a s y m t o t i c a l l y a x 2 d i s t r i b u t i o n with the degrees of freedom equal to the number of parameters (31 i n t h i s case) estimated. The x 2 t e s t s t a t i s t i c o b t ained f o r the t r i a l sample was 452.1 with 31 degrees of freedom. T h i s r e j e c t e d the hypothesis that ownership p r o b a b i l i t i e s vary only randomly a c r o s s f a m i l i e s . Measures of the model's " g o o d n e s s - o f - f i t " , as e x p l a i n e d i n Chapter I I , are the "pseudo-R 2" or and r e g u l a r R 2. It appeared that the estimated c o e f f i c i e n t s provide a modest f i t to the data. The pseudo-R 2 and R 2 are 0.218 and 0.268 r e s p e c t i v e l y . F u r t h e r , the magnitude of the estimated parameters f o r the r e p l i c a t i o n sample and the t o t a l sample were comparable. A d e t a i l e d d e s c r i p t i o n of the model based on the t o t a l sample i s p r o v i d e d i n the remainder of t h i s s u b s e c t i o n . Confirmatory model P r e d i c t e d p r o b a b i l i t i e s of the number of c a r s owned by a f a m i l y can be obtained by i n s e r t i n g values f o r the v a r i a b l e s i n t o the estimated model as shown in Table 4.1. For example, an average Canadian f a m i l y earned about 18 thousand d o l l a r s i n 1978 and c o n s i s t e d of three or four members. They l i v e d i n a c i t y with a p o p u l a t i o n of 400,000 and were r e s i d e n t s of e i t h e r O n t a r i o or Quebec. In terms of l i f e - c y c l e , the modal famil y type was a young f a m i l y with a c h i l d a t t e n d i n g school and i f another c h i l d was present, that c h i l d was at the l e v e l of p r e - s c h o o l . According to estimated parameters, such an average f a m i l y had 54.2%, 34.3% and 11.5% chances of owning one, two and three or more c a r s r e s p e c t i v e l y . The i n f l u e n c e of i n d i v i d u a l 94 v a r i a b l e s on these percentages are d i s c u s s e d below. There was no s u r p r i s e i n observing that income and household s i z e both had a p o s i t i v e impact (p<0-0l) on the number of c a r s owned by the household. More s p e c i f i c a l l y , l o o k i n g at a marginal i n c r e a s e i n income by a standard d e v i a t i o n (about 9 thousand) from the sample average i n c r e a s e d the r e l a t i v e p r o b a b i l i t i e s of owning two c a r s from 34.3% to 43.97%. However, the e f f e c t of a s i m i l a r increase i n income i n c r e a s e d the c h o i c e p r o b a b i l i t i e s of the f a m i l y owning three c a r s by 5.29%. T h i s i m p l i e s that income does not d i s c r i m i n a t e w e l l between owners of two, three or more c a r s . On the other hand, the e f f e c t of f a m i l y s i z e on the c h o i c e p r o b a b i l i t i e s was weaker for two compared to one car owners than i t was f o r three compared to two c a r s owners. For example, an i n c r e a s e i n f a m i l y s i z e from three to f i v e persons, decreased the chances of a f a m i l y owning one car from 57.8% to 39.1% and that of owning three c a r s from 8.9% to 26.1%. Although the r e s u l t was not e n t i r e l y s u r p r i s i n g , i t d i d suggest that l a r g e r households r e q u i r i n g g r e a t e r f l e x i b i l i t y i n t r a n s p o r t a t i o n , tend to own more v e h i c l e s . Because past r e s e a r c h had i n d i c a t e d the importance of a f a m i l y ' s employment s t a t u s on car ownership, i t was d e c i d e d to i n v e s t i g a t e the impact of the wife's employment s t a t u s on the number of c a r s owned. As might be expected, t h i s v a r i a b l e p o s i t i v e l y , though not strongly(p<0•10), a f f e c t e d the number of c a r s owned. The estimated parameters suggested that f a m i l i e s 95 with an employed wife had a 51.6% chance of owning one car as opposed to a 55.9% chance f o r f a m i l i e s without a wife working. I t was noted i n chapter II that the age of the f a m i l y head d i d not d i s c r i m i n a t e between two and no car owners. Hence, a c a r e f u l l i f e - c y c l e c l a s s i f i c a t i o n ( T a b l e 4.2), based on the work of Murphy and Staples(1979) was used to i d e n t i f y n o n - l i n e a r e f f e c t s of age on the number of v e h i c l e s owned by a f a m i l y . V a r i o u s c l a s s i f i c a t i o n v a r i a b l e s showed s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e s from " f i n a n c i a l l y c o n s t r a i n e d groups" ( s i n g l e p a r e n t s , r e t i r e d , and middle age f a m i l i e s with a youngest c h i l d l e s s than four years o l d ) , and these d i f f e r e n c e s a c r o s s l i f e - c y c l e groups appeared systematic. S e v e r a l key o b s e r v a t i o n s from the estimated c o e f f i c i e n t are summarized below. The f i n a n c i a l l y c o n s t r a i n e d group had a 69.82% chance of owning one v e h i c l e , and the l e a s t chance (23.94%) among the e n t i r e sample of owning two c a r s . For f a m i l i e s without c h i l d r e n , an i n c r e a s e i n age, i n c r e a s e d the chances of owning two c a r s , and the chances of owning three c a r s appeared convex ( i n v e r t e d "U"). For middle age f a m i l i e s with the age of the youngest c h i l d g r e a t e r than four, i n c r e a s e s i n the age of the youngest c h i l d decreased the chances of a household owning one car and i n c r e a s e d the chance of owning m u l t i p l e c a r s . Thus, middle age f a m i l i e s with c h i l d r e n above 12 years o l d had 20.86%, 38.03%, and 41.11% chances of owning one, two, and three c a r ( s ) r e s p e c t i v e l y . "Older" f a m i l i e s with employed f a m i l y heads or 9<2> T a b l e 4.2 Impact of L i f e - C y c l e on Number of Cars Owned Age of C h i l d Youngest Two P a r e n t Households Husband's age young <35 mi d d l e 35-65 o l d >65 S i n g l e P a r e n t Households No c h i l d r e n < 4 y e a r s 4 - 8 y e a r s 8 - 1 2 y e a r s > 12 y e a r s p,=46.88 p 2=41.56 p 3=11.56 p,=46.22 p 2=32.85 p 3=20.94 p,=59.74 p 2=35.25 p 3= 5.01 p,=69.82 p 2 = 23 .94 p 3 = 6.24 p,=48.74 p 2=44.48 p 3= 6.77 p,=53.59 p 2 = 37.33 p 3= 9.08 p!=45.27 p 2=36.83 p 3=17.90 p,=20.86 p 2=38.03 p 3 = 41 . 1 0 P!=69.82 p 2=23.94 p 3= 6.24 p,=44.96 p 2=34.72 p 3 = 20 . 33 P!=69.82 p 2=23.94 p 3= 6.24 p,, p 2 and p 3 are p r o b a b i l i t i e s ( e x p r e s s e d i n p e r c e n t ) of one, two and t h r e e or more c a r s r e s p e c t i v e l y . Note t h a t each c e l l i n the above m a t r i x r e p r e s e n t (0 or 1 ) i n d i c a t o r v a r i a b l e i n ML e s t i m a t i o n . 97 with an a d d i t i o n a l f a m i l y member l i v i n g with them, had a b e t t e r chance of owning m u l t i p l e c a r s than were o l d e r r e t i r e d c o u p l e s . Although g a s o l i n e p r i c e s i n the province of A l b e r t a have been the lowest i n Canada for the l a s t ten years, the p r e d i c t e d ownership p r o b a b i l i t i e s across a l l of the western p r o v i n c e s showed s i m i l a r p a t t e r n s . F u r t h e r , f a m i l i e s from these p r o v i n c e s tended to own more v e h i c l e s per household than f a m i l i e s from ea s t e r n Canada. Observed and p r e d i c t e d p r o b a b i l i t i e s f o r two car owners acr o s s a l l p r o v i n c e s and average g a s o l i n e p r i c e s i n major c i t i e s w i t h i n each of the p r o v i n c e s (Table 4.3) showed a modest negative a s s o c i a t i o n ( r = - 0 . 5 6 ) with ownership. Thus, there was evidence at the aggregate l e v e l that the consumer d e c i s i o n about the number of c a r s owned was a f f e c t e d by g a s o l i n e p r i c e s . Moreover, estimated parameters for the eastern region suggested that Quebec f a m i l i e s own the l e a s t number of c a r s per f a m i l y , f o l l o w e d by f a m i l i e s i n the A t l a n t i c p r o v i n c e s and O n t a r i o . A l l these d i f f e r e n c e s were s t a t i s t i c a l l y s i g n i f i c a n t at p<0-00l. As might be expected, the p o p u l a t i o n of the c i t y i n which the f a m i l y r e s i d e d had a negative e f f e c t (p^0«00l) on the number of c a r s owned by the f a m i l y . More s p e c i f i c a l l y , those l i v i n g i n a c i t y with a p o p u l a t i o n of 400,000 had 54.2% and 34.3% chance of owning one and two c a r s r e s p e c t i v e l y . However, those l i v i n g i n a l a r g e m e t r o p o l i t a n c e n t r e (Toronto, Montreal and Vancouver), had 64.92% and 27.43% chance of owning one and two c a r s r e s p e c t i v e l y . Thus, i t may be that the a v a i l a b i l i t y of T a b l e 4.3 Impact of G a s o l i n e P r i c e s on Number of Cars Owned 9 & Observed G a s o l i n e p r i c e s * P r o v i n c e p r o p o r t i o n on Dec.1, on March 1, i n C i t y of two c a r 1977 1979 own i ng fami l i e s ( i n %) Newfoundland 21 .43 1 07 .9 115.4 S t . Johns P r i n c e Edward I s l a n d 41 .67 1 02 .0 109.9 C h a r l o t t e t o w n Nova S c o t i a 32.35 97 . 4 1 02.6 H a l i fax New Brunswick 41 .24 94 .9 99.9 S a i n t John Quebec 24 . 56 85 .9 97 .7 M o n t r e a l Ontar i o 43 . 07 86 . 9 99.9 Toronto Man i toba 38 .33 90 .9 96.7 Winnipeg Saskatchewan 48.31 88 . 9 99.0 Reg i na A l b e r t a 49. 70 81 .9 80.8 Edmonton B r i t i s h Columbia 52 . 02 86 .9 94.9 Vancouver Region P r e d i c t e d p r o p o r t i o n s f o r one two and t h r e e c a r owning f a m i l i e s ( i n %) Weighted p r i c e of on March average g a s o l i n e 1, 1979+ A t l a n t i c 58. 44 31.61 9.95 1 06. 0 Quebec 75.84 20.05 4.11 97. 7 Ontar i o 49. 99 36. 65 13.36 99. 9 Western 35.17 43. 63 2 1.20 92. 8 * G a s o l i n e p r i c e s o b t a i n e d from O i l w e e k , v o l . 2 9 ( 9 ) , A p r i l 10, 1978, p.18 and O i l w e e k , v o l . 3 0 ( 1 9 ) , June 18, 1979, p.16. + Weighted average g a s o l i n e p r i c e c a l c u l a t e d by w e i g h i n g the g a s o l i n e p r i c e of each p r o v i n c e by the number of two c a r owning f a m i l i e s . Note: C o r r e l a t i o n between the p r o p o r t i o n of two c a r owning f a m i l i e s and g a s o l i n e p r i c e s i n 1977 was -0.555. The same c o r r e l a t i o n i n 1979 was -0.596. 99 p u b l i c t r a n s p o r t a t i o n reduces the need f o r a d d i t i o n a l automobiles. The e f f i c a c y of the model summarized above with a l a r g e sample, compares w e l l with the r e s u l t s of past econometric r e s e a r c h . Income, f a m i l y s i z e and employment s t a t u s , a l l p o s i t i v e l y a f f e c t the number of c a r s owned by a f a m i l y . Although the past research has t e s t e d l i n e a r e f f e c t r e l a t e d to age of fami l y head and f a i l e d to provide c o n s i s t e n t r e s u l t s , t h i s study p r o v i d e s c o n s i s t e n t although, n o n - l i n e a r e f f e c t s of age. Regional v a r i a b l e s , as p r o x i e s f o r g a s o l i n e p r i c e s , d i d pr o v i d e p a r t i a l evidence to support the hypothesis that g a s o l i n e p r i c e s n e g a t i v e l y a f f e c t s the number of c a r s owned. It was a l s o concluded that f a m i l i e s with access to p u b l i c t r a n s p o r t a t i o n tended to own fewer automobiles. E x p l o r a t o r y Model The model d e s c r i b e d above was based on a small subset of v a r i a b l e s , a n d . i t provided a modest f i t to the sample o b s e r v a t i o n s . E x c l u s i o n of important v a r i a b l e s i n the model, however, c o u l d provide b i a s e d estimates of the remaining parameters. The c o s t s of i n c l u d i n g a l a r g e set of v a r i a b l e s , however, i s the l o s s of c o m p a r a b i l i t y of the estimated parameters with those of other s t u d i e s as w e l l as a r e d u c t i o n i n sample s i z e due to missing o b s e r v a t i o n s . Hence, the model d e s c r i b e d below must be i n t e r p r e t e d with c a u t i o n . The t r i a l sample a n a l y s i s that u t i l i z e d a t t i t u d e r e l a t e d f a c t o r s suggested that of the 14 v a r i a b l e s chosen f o r a n a l y s i s , p o s i t i v e d i s p o s i t i o n towards l a r g e c a r s and u n w i l l i n g n e s s to pay 100 higher g a s o l i n e p r i c e s were a s s o c i a t e d with f a m i l i e s owning a l a r g e r numbers of c a r s . F u r t h e r , consumer p a r t i c i p a t i o n i n a c t i v i t i e s such as keeping record of g a s o l i n e expenses were common to two car owner f a m i l i e s . The model that t e s t e d t r a v e l need r e l a t e d v a r i a b l e s suggested that t o t a l weekly commuting to work and need of a car at work, were p o s i t i v e l y a s s o c i a t e d with the l e v e l of car ownership. I t was a l s o r e v e a l e d that the " p r o p o r t i o n of d r i v i n g on c i t y roads", had a negative impact on the number of cars owned, but i t d i d not o f f e r b e t t e r d i s c r i m i n a t i o n between the v a r i o u s l e v e l s of car ownership. The f i n a l e x p l o r a t o r y model proposed to p r e d i c t the p r o b a b i l i t i e s of the number of cars owned c o n s i s t e d of income, fa m i l y s i z e , the l i f e - c y c l e p o s i t i o n of fa m i l y , r e g i o n , p o p u l a t i o n of c i t y of r e s i d e n c e , t r a v e l needs, a measure of a t t i t u d e toward e f f i c i e n t c a r s , and an energy c o n s e r v a t i o n r e l a t e d measure. Maximum l i k e l i h o o d parameter estimates and t h e i r t - s t a t i s t i c s f o r these v a r i a b l e s are presented i n Table 4.4. In the remainder of t h i s s u b s e c t i o n , the i m p l i c a t i o n s of t r a n s p o r t a t i o n needs and a t t i t u d i n a l v a r i a b l e s on p r o b a b i l i t y estimates from the t o t a l sample are d i s c u s s e d . On the average, a Canadian f a m i l y commuted 90 m i l e s per week to work and at t h i s l e v e l of commuting had 57.99%, 31.4, and 10.7% chances of owning one, two, and three c a r ( s ) r e s p e c t i v e l y . However, a one standard d e v i a t i o n i n c r e a s e i n commuting (128 miles) decreased the chance of owning one car to 101 Table 4.4 MLE f o r P r e d i c t i n g P r o b a b i l i t i e s of Number of Cars Owned S e l e c t e d set of V a r i a b l e s ( t - s t a t i s t i c s i n parentheses) A l l t - s t a t i s t i c s l e s s than one i n a b s o l u t e value omitted. Variables T r i a l sample consumer choice Two(0) vs. One(1) car{s) Two(0) vs. Three(1) cars R e p l i c a t i o n sample Consumer choice Two(0) vs. One(l) car(s) Two(0) vs. Three(1) cars Total sample Consumer choice Two(0) vs. One(1) car(s) Two(0) vs. Three(1) cars Income (in '000) Household s i z e If wife i s employed (0=No; l=yes) Young couple c h i l d r e n No Young family with youngest c h i l d below 4 years Young couple with youngest c h i l d above 4 years Middle age family with no c h i l d r e n Middle age family with youngest c h i l d between 4 to 8 yrs Middle age family with youngest c h i l d between 8 to 12 yrs Middle age family with youngest c h i l d above 12 years Old age family with some other family member l i v i n g A t l a n t i c region Quebec Ontario C i t y s i z e (i n '000) Total weekly commuting to work If car(s) used at work (0=No; l=Yes) Manufactures should not be allowed to produce cars with l e s s than 25 MPG (l=disagree; 6=agree) Keep record of gas expenses (l=Yes; 2=N0) Constant -0.0345 (-2.50) -0.139 (-1.25) -0.237 (-1.04) -0.0427 -0.195 -0.473 -0.630 (-1.35) -0.959 (-2.18) -0.714 (-1.61) -1.648 (-3.56) -0.497 1.120 (2.84) 1.538 (5.20) 0.438 (1.79) 0.000760 (2.89) -0.00594 (-5.65) -0.596 (-2.23) 0 . 264 (3.92) 0 . 266 (1.32) 1 .482 (2.42) 0.0407 (1.53) 0.941 (4-46) 0.211 1.002 -0.108 -0.948 1 . 584 (1.25) -2.772 (-2.04) 1 . 240 (1.32) 1 . 873 (1.85) 2.356 (1.99) -0.547 0.623 (1.40) 0.000542 (1.04) 0.00140 (1.10) 0.207 0.0665 -0.464 (-1.01) -2.654 -0.0795 (-5.34) -0.0185 -0.171 -0.935 (-1.61) -0.444 -0.579 0.415 -1 .048 (-2.20) -1.090 (-2.33) -1.788 (-3.22) -0.663 (-1.13) 0 . 633 (1.58) 1.673 (5.09) 0 . 329 (1.25) 0.000574 (2.09) -0.00639 (-5.15) -0.871 (-3.02) 0.341 (1.56) 1 . 463 (2.39) 0.00294 0.622 (2.37) -0.714 (-1.58) 0.0 -2.618 (-2.09) -0.496 1 .275 (1.12) -1.059 0.487 1 .974 (2.42) 0.776 -0.968 (-1.42) -1.952 (-2.36) -1.407 (-2.92) -0.00074 (-1.34) 0.00155 (1.14) 0. 177 -0.212 (-1.73) -0.0634 -1.480 -0.0549 (-5.57) -0.0928 (-1.17) -0.187 (-1.17) -0.454 (-1.20) -0.258 -0.413 (-1.08) -0.259 -0.976 (-2.08) -0.864 (-2.76) -1.662 (-4.75) -0.577 (-1.49) 0.858 (3.19) 1.572 (7.30) 0.394' (2.24) 0.000651 (3.51) -0.00606 (-7.73) -0.699 (-3.19) 0. 115 (2.48) 0 . 254 (1.76) 1. 501 (3.67) 0.0188 (1.11) 0.731 (5.39) -0.262 -0.117 -1.268 (-1.70) -0.492 1 . 114 (1.34) -1.430 (-2.08) 0.832 (1.42) 1 .658 (2.64) 1 .398 (1.82) -0.712 (-1.23) -1.389 (-2.70) -0.403 (-1.33) -0.00018 0.00116 (1.37) 0.258 -0.0727 -0.346 (-1.27) -2.523 Pseudo R2 R* l i k e l i h o o d r a t i o test cross v a l i d a t e d Ra 0.340 0.403 406.56 with 38 df 0 . 326 0. 352 0.415 393.4 with 38 df 0.318 742 0.321 0.388 ,1 with 39 df Sample d i s t r i b u t i o n s one car owners two car owners three car owners Total 361 230 44 635 329 220 41 590 690 450 86 1225 103 38.22% and in c r e a s e d the chance of owning two and three cars to 44.34% and 17.44% r e s p e c t i v e l y . Although d i s c r i m i n a t i o n between two and one car owners was s t a t i s t i c a l l y s i g n i f i c a n t (p<0*00l), d i s c r i m i n a t i o n between two and three car owners was not. T h i s i m p l i e d , a d e c l i n i n g marginal impact of t o t a l commuting to work on number of c a r s owned. Although only 18.4% of the sample f a m i l i e s used c a r ( s ) at work, these f a m i l i e s r e q u i r e d a gre a t e r number of c a r s . S p e c i f i c a l l y , f a m i l i e s using c a r s at work had 61.3%, 29.2, and 9.5% chance of owning one, two, and three c a r ( s ) r e s p e c t i v e l y and those not needing a car at work had a 42.32%, 40.62, and 17.1% chance of owning one, two, and three c a r ( s ) r e s p e c t i v e l y . These r e s u l t s were s t a t i s t i c a l l y s i g n i f i c a n t at p^0«0l. In 1978, Canadian consumers g e n e r a l l y approved of the i n d u s t r y manufacturing c a r s with more than 25 MPG. However, the gr e a t e r the approval r a t i n g , fewer c a r s were owned by a f a m i l y . Thus, i n c r e a s i n g the pr e f e r e n c e f o r e f f i c i e n t v e h i c l e s by two u n i t s from the n e u t r a l p o i n t on the r a t i n g s c a l e ( 3 ) , decreased the chances of a f a m i l y owning two c a r s from 34.75% to 31.15% and i n c r e a s e d the chances of owning one car., from 51.7% to 58.34%. T h i s i m p l i e d that consumers p r e f e r r i n g f u e l e f f i c i e n t v e h i c l e s owned fewer v e h i c l e s . Keeping a r e c o r d of g a s o l i n e consumption was an i n d i c a t i o n of " t h r i f t i n e s s " , and the s t a t i s t i c a l r e s u l t s suggested that t h r i f t y consumers owned more automobiles. For example, record keeping consumers had 52.91%, 33.24, and 13.85 chances of owning 1 04 one, two and three c a r ( s ) r e s p e c t i v e l y . On the other hand, non-record keeping consumers had 61.31%, 29.88%, and 8.81% chances for the same. T h i s suggested, consumers owning more than one car kept records of g a s o l i n e consumption. Most of the remaining c o e f f i c i e n t s , even with a l o s s of about two - t h i r d s of the sample, showed s i m i l a r i t y between the models r e p o r t e d i n Table 4.1 and Table 4.4. The most noteworthy d i f f e r e n c e was not i n the d i r e c t i o n of e f f e c t but i n the magnitude of l i f e - c y c l e i n d i c a t o r s . However, the two models summarized in t h i s s e c t i o n , p r o v i d e d strong evidence that a consumer's c h o i c e of the number of c a r s owned i s the r e s u l t of systematic c h o i c e behaviour. 4.3 Models P r e d i c t i n g Average Car Weight Although there have been p r i o r attempts to understand consumer c h o i c e s of automobile s i z e , past r e s e a r c h has not been systematic enough to allow u s e f u l i m p l i c a t i o n s to be drawn about energy consumption. The purpose of t h i s s e c t i o n , i s to summarize v a r i o u s models that were c o n s t r u c t e d to p r e d i c t c h o i c e of average car weight w i t h i n a f a m i l y . In the preceding s e c t i o n , i t was noted that consumer demographic and s i t u a t i o n a l f a c t o r s were the s t r o n g e s t determinants of the number of c a r s owned. R e s u l t s reported i n t h i s s e c t i o n , however, suggested that a consumer's choice about average car weight was p r i m a r i l y i n f l u e n c e d by a t t i t u d e and t r a v e l need r e l a t e d f a c t o r s . 1 05 T r i a l sample The model chosen f o r the t r i a l sample was based on r e g r e s s i o n a n a l y s i s (stepwise as w e l l as a l l - s u b s e t ) using four s e t s of independent v a r i a b l e s . The f i r s t set c o n s i s t e d of demographic v a r i a b l e s such as income and household s i z e while the second set c o n s i s t e d of r e g i o n a l v a r i a b l e s such as degree-days and the s i z e of the urban c e n t r e . The t h i r d set of independent v a r i a b l e s i n c l u d e d t r a v e l need r e l a t e d f a c t o r s , and the f o u r t h set c o n s i s t e d of energy r e l a t e d a t t i t u d e s and p a r t i c i p a t i o n items. The "best model" for each set of v a r i a b l e s i s summarized in Table 4.5. The p r e d i c t o r set that, i n c l u d e d demographic v a r i a b l e s i n d i c a t e d f i v e s i g n i f i c a n t v a r i a b l e s at p<0«l0. Family s i z e and middle age couple v a r i a b l e s were both p o s i t i v e l y r e l a t e d to average automobile weight. F u r t h e r , the wife's employment s t a t u s , as w e l l as the husband's education and occupation were n e g a t i v e l y a s s o c i a t e d with average car weight. On the whole, these f i v e f a c t o r s accounted f o r only 5.8% of the t o t a l v a r i a t i o n . The r e g r e s s i o n a n a l y s i s i n v o l v i n g r e g i o n a l v a r i a b l e s as w e l l as c l i m a t i c f a c t o r s i n d i c a t e d that severe wi n t e r s , l a r g e c i t y r e s i d e n c y , and non-residency i n B r i t i s h Columbia, i n c r e a s e d the average automobile weight. F u r t h e r , these three i n d i c a t o r s by themselves accounted f o r only 3.6 of the v a r i a n c e . Although the economic l i f e - c y c l e theory in chapter I I , suggested that the consumer's c h o i c e of a house and the l o c a t i o n should be r e l a t e d to automobile c h o i c e , the e m p i r i c a l evidence does not support . Table 4.5 T r i a l Sample Models of Average Car Weight" (standardized beta c o e f i c i e n t s i n parentheses) Demograprephic v a r i a b l e set S i t u a t i o n a l v a r i a b l e set Car r e l a t e d v a r i a b l e set Attitu d e and P a r t i c i p a t i o n v a r i a b l e set Wife's employment status (-0.112) l=yes; 0=no Degree-days(0.112) Proportion of cars with a i r - c o n d i t i o n i n g (0.259) Large car owners should be taxed $300 per year (-0.268) Husband's education low=l; High=4 (-0.126) If l i v e i n large c i t y 0=No; l=yes (-0.104) Number of cars owned (-0.189) Compact car would serve my needs (-0.207) s i g n i f p<0 .1 Family s i z e (0.105) If resident of B.C.(-0.0745) If own one or more re c r e a t i o n a l (0.144) Cars should have at l e a s t one passenger(0.097) If middle age couple (0.13B) F l e e t ' s oldness (0.138) Present car smaller than the previous (0.237) Husband 1s occupation(-0.079) l = o f f i c e r e l a t e d ; 0=labour oriented. Car was tuned-up twice l a s t year (0.110) Usually take bus to work(0.080) R2 0.058 0.0362 0.101 0.244 Not s i g n i f . p<0.1 Income. Other l i f e - c y c l e groups. If family has more than 4 members. Other regions. C i t y ' s population s i z e . Type of house. If car used at work. Total weekly commuting. Proportion d r i v i n g i n c i t y . Remaining eight a t t i t u d e and p a r t i c i p a t i o n items from Table 3.5 "Weight measured i n lOOlbs. 1 07 t h i s n o t i o n for the Canadian sample. The r e g r e s s i o n a n a l y s i s i n v o l v i n g t r a v e l need v a r i a b l e s i n d i c a t e d that fewer c a r s , c a r s with a i r - c o n d i t i o n i n g , ownership of a c a m p e r / t r a i l e r / b o a t and an o l d e r (on average) automobile f l e e t , were a s s o c i a t e d with a higher average f l e e t weight. Although these four f a c t o r s d i d b e t t e r than the two previous v a r i a b l e subsets, t h i s subset only accounted f o r 10% of the v a r i a t i o n of f l e e t weight ac r o s s the sample f a m i l i e s . Moreover, i t d i d not appear that consumers with g r e a t e r commuting requirements d e s i r e d g r e a t e r comfort and thus, bought bigger c a r s . Of the 14 responses to a t t i t u d e and p a r t i c i p a t i o n items, s i x items showed a s i g n i f i c a n t a s s o c i a t i o n at P<0«10 and accounted f o r 24% of the v a r i a t i o n in average car weight. S p e c i f i c a l l y , those c l a i m i n g that they purchased smaller c a r s , tuned-up t h e i r c a r s and rode the bus to work, tended to own s m a l l e r c a r s . F u r t h e r , those p r e f e r r i n g compact c a r s , or those who would l i k e to tax l a r g e c a r s or would be u n w i l l i n g to have an a d d i t i o n a l passenger in t h e i r c a r , tended to own smaller c a r s . I t was expected that consumers p r e f e r r i n g new c a r s , or lower g a s o l i n e p r i c e s , or p r e f e r e n c e f o r l a r g e c a r s would a l s o own l a r g e c a r s . The e m p i r i c a l evidence d i d not support these e x p e c t a t i o n s . In the c o n s t r u c t i o n of the f i n a l model, a l l independent v a r i a b l e s were used. In Table 4.6, the r e s u l t s of v a r i a b l e s e l e c t i o n and parameter e s t i m a t i o n f o r the t r i a l , as w e l l as 108 Table 4.6 Models P r e d i c t i n g Average Car Weight 1 ( t - s t a t i s t i c s i n parentheses) 1weight measured i n lOOlbs. Regression c o e f f i c e n t s f o r T r i a l R e p l i c a t ion . T o t a l V a r i a b l e s sample sample sample Demographic f a c t o r s Household s i z e 0.461 0.822 0.671 (2.33) (4.06) (4.77) If wife employed -1.826 -0.983 -1.426 (0=No; 1=Yes) (-3.92) (-1.89) (-4.14) If middle age 1 .408 1 .070 1 .246 couple (1.89) (1.28) (2.24) S i t u a t i o n a l f a c t o r s If l i v e i n l a r g e -1.529 -0.224 -0,910 c i t y (-2.53) (-0.33) (-2.02) (0=No; 1=Yes) Degree-day 0.000324 0.00039 0.000353 (degrees F) (2.08) (2.19) (3.03) Car r e l a t e d f a c t o r s Number of c a r s -2.290 -3.361 -2.816 owned by fami l y (-5.53) (-7.13) (-9.08) P r o p o r t i o n of 2.730 4. 1 94 3.479 c a r s with (4.06) (6.67) (8.13) a i r - c o n d i t i o n i n g (0=None; 1=A11) If one or more 0.606 0.522 0.568 r e c r e a t i o n a l (2.16) (1.81) (2.84) items owned (0=No; 1=Yes) F l e e t s oldness 0. 1 73 0.246 0.212 ( i n years) (2.01 ) (2.78) (3.45) P a r t i c i p a t i o n and Ati :itude f a c t o r s Present car 2.689 3.486 3.099 small e r than the (5.67) (6.88) (8.98) l a s t . (1=Yes; 2=No) Compact car would -0.692 -0.545 -0.621 serve my needs (-5.30) (-4.00) (-6.61) (1=di sagree; 6=agree) Should tax $300 -0.830 -0.732 -0.762 on l a r g e c a r s . (-6.21) (-5.21) (-7.91) (1=disagree; 6=agree) Constant 35.050 31.131 32.956 (17.25) (13.89) (22.13) R 2 a d j 0.325 0.333 0.329 c r o s s - v a l i d a t i o n R 2 0.318 0.315 Sample s i z e 585 533 1118 1 10 r e p l i c a t i o n and t o t a l sample are presented. The t r i a l sample model parameters accounted f o r 32.5% of v a r i a t i o n a c r o s s f a m i l i e s and provided about the same l e v e l of c r o s s - v a l i d a t i o n . S i m i l a r l e v e l s of e x p l a n a t i o n were achieved with the r e p l i c a t i o n sample. The d e s c r i p t i o n of model parameters based on the t o t a l sample i s presented i n the remainder of t h i s s u b s e c t i o n . Average Car Weight Model On an average, a Canadian f a m i l y owned an automobile of 34601bs in 1978 which achieved 18.7 MPG on c i t y roads and highways. I t was suggested i n chapter I, that a 45 l b s change i n an automobile's weight r e s u l t s i n approximately one percent change in g a s o l i n e consumption. Thus, the model p r e d i c t i n g automobile weight p r o v i d e s an i n d i c a t i o n of the change in g a s o l i n e consumption that may be expected with changes in average car weight. T h i s i n f o r m a t i o n about g a s o l i n e saving i s p r ovided i n parentheses i n the f o l l o w i n g d i s c u s s i o n . In terms of demographic f a c t o r s , f o r each a d d i t i o n a l f a m i l y member, the household tended to own c a r ( s ) that weighed an a d d i t i o n a l 67 lbs(1.5%) each. F u r t h e r , f a m i l i e s with an employed wife tended to own an automobile f l e e t s maller by 142 l b s ( 3 . 1 6 % ) , while middle age couples owned f l e e t s l a r g e r by 125 lbs(2.78%) as compared to an average f a m i l y . In an e a r l i e r s e c t i o n , i t was noted that f a m i l i e s with an employed wife, tended to own m u l t i p l e v e h i c l e s . The car weight model suggested that these f a m i l i e s chose to own two mid-sized or one l a r g e and one s m a l l - s i z e d c a r . On the other hand, a middle age couple appeared l e s s l i k e l y to own two c a r s , and p r e f e r r e d to own a 111 s i n g l e l a r g e r v e h i c l e . Two s i t u a t i o n a l f a c t o r s , r e s i d e n c y i n l a r g e c i t i e s and severeness of w i n t e r s , both a f f e c t e d average car weight n e g a t i v e l y . For example, consumers from r u r a l r e g i o n s , tended to own automobiles which were 91 l b s ( 2 % ) l a r g e r than those owned by consumers l i v i n g in urban r e g i o n s . F u r t h e r , consumers l i v i n g in areas with 5500°F annual degree-days had c a r s smaller by 100 lbs ( 2 . 2 % ) than an average Canadian l i v i n g i n a 8260°F annual degree-days area. Both of these e f f e c t s were expected and were s i g n i f i c a n t at p<0«0l. F a m i l i e s owning more ca r s have ch o i c e s of owning d i f f e r e n t s i z e s of c a r s . The e m p i r i c a l evidence supported the p o s s i b i l i t y that m u l t i p l e car consumers bought smaller c a r s to reduce g a s o l i n e consumption. The estimated model parameters suggested that f a m i l i e s owning m u l t i p l e v e h i c l e s reduces the average car weight per v e h i c l e by 282 lbs(6.27%) f o r every a d d i t i o n a l v e h i c l e . T h i s i n turn suggests, that f a m i l i e s may attempt to reduce g a s o l i n e consumption by matching t r i p purpose with v e h i c l e s i z e . In the ECCP sample,. 20.4% of Canadian f a m i l i e s had one or more c a r s which were a i r - c o n d i t i o n e d , and these c a r s weighed 348 lbs(7.73%) more than v e h i c l e s without a i r - c o n d i t i o n i n g . S i m i l a r l y , 66.8% of the sample f a m i l i e s owned one or more r e c r e a t i o n a l items, and these f a m i l i e s r e q u i r e d t h e i r c a r s to be 57 lbs(1.27%) h e a v i e r . I t was not s u r p r i s i n g to f i n d that o l d e r v e h i c l e s tended to be h e a v i e r . On an average, the estimated 1 1 2 parameter i n d i c a t e d that average car weight decreased by 21 lbs(0.47%) per year in 1978. Although t h i s d e c l i n e seemed to be s m a l l e r than the new v e h i c l e f l e e t weight from 1978 to 1982 (Consumer Reports, 1978-1982), the c o e f f i c i e n t of 21lbs provided an i n d i c a t i o n of the r e l a t i v e slowness in the automobile stock adjustment p r o c e s s . It was hypothesised that energy c o n s e r v a t i o n a t t i t u d e s and p a r t i c i p a t i o n would d i r e c t l y i n f l u e n c e a consumer's c h o i c e of automobile s i z e . The r e s u l t s suggested that consumers c l a i m i n g that they own smaller c a r s tended to own c a r s s m a l l e r by 310 l b s ( 6 . 8 9 % ) . F u r t h e r , consumers w i l l i n g to s a t i s f y t h e i r t r a v e l needs with compact c a r s , owned smal l e r c a r s by 311 l b s ( 6 . 9 l % ) compared to u n w i l l i n g consumers. S i m i l a r l y , consumers w i l l i n g to pay $300 annual tax on l a r g e c a r s , owned smal l e r c a r s by 381 l b s ( 8 . 4 7 % ) . F u r t h e r , i f the a d d i t i v e e f f e c t s of these three f a c t o r s were co n s i d e r e d together, the net e f f e c t of g a s o l i n e c o n s e r v a t i o n was about 22.25% over the average g a s o l i n e consumption p a t t e r n . To summarize, the automobile f l e e t weight model suggested that the c h o i c e of an automobile was s y s t e m a t i c . F u r t h e r , the c h o i c e depended on demographic, s i t u a t i o n a l , t r a v e l need r e l a t e d , and a t t i t u d i n a l f a c t o r s . I t appeared from the e m p i r i c a l a n a l y s i s that a t t i t u d e s towards l a r g e c a r s , a consumer's c l a i m of involvement in energy c o n s e r v a t i o n a c t i v i t i e s , and t r a v e l need r e l a t e d f a c t o r s were the most important p r e d i c t o r s of average automobile weight. 1 1 3 4.4 Comprehensive Model of Consumer Car F l e e t Choice The two models d e s c r i b e d i n previous s e c t i o n s provide a view of consumer automobile f l e e t c h o i c e s . These two models provide a reasonable f i t to the data, and t h e i r d e s c r i p t i o n seems to be c o n s i s t e n t with past econometric and marketing resea r c h f i n d i n g s . In t h i s s e c t i o n , an attempt i s made to i n t e g r a t e both models and provide a comprehensive view of the c a u s a l r e l a t i o n s h i p s between household f l e e t d e c i s i o n s and demographic, s i t u a t i o n a l , t r a v e l need, and a t t i t u d e r e l a t e d f a c t o r s (Figure 4.1). Demographic f a c t o r s As might be expected, income a f f e c t s the number of c a r s owned by a f a m i l y . However, i t s impact on average car weight i s not s t a t i s t i c a l l y s i g n i f i c a n t . Family s i z e p o s i t i v e l y a f f e c t s both number and weight. The e f f e c t of l i f e - c y c l e stages i s n o n - l i n e a r f o r the number of c a r s owned. More s p e c i f i c a l l y , consumers appear to own the number of c a r s based on t h e i r l i f e - c y c l e needs. The wife's employment s t a t u s p o s i t i v e l y a f f e c t s the number of c a r s owned, but n e g a t i v e l y a f f e c t s the average car s i z e . T h i s i m p l i e s that consumers r e q u i r i n g g r e a t e r d r i v i n g f l e x i b i l i t y achieve i t through owning m u l t i p l e s m a l l e r v e h i c l e s . S i t u a t i o n a l f a c t o r s Residency i n l a r g e c i t i e s and g a s o l i n e p r i c e s a f f e c t both n e g a t i v e l y the number of c a r s owned and the average car weight. I t i s suspected in t h i s study, that the g a s o l i n e p r i c e s may a f f e c t average car weight but t h i s may be P a r t i c i p a t i o n i n g a s o l i n e c o n s e r v a t i o n •bought a compact c a r P e r c e i v e d c o s t of s i z e p r e f e r e n c e s •Tax $300 f o r b i g a u t o m o b i l e s S i z e p r e f e r e n c e s •compact c a r s e r v e s my need R e c r e a t i o n a l i t e m ownership e.g. camper, t r a i l e r c-o c-O i i si C+-) F l e e t o l d n e s s • average age of f l e e t F l e e t equipment • c a r s w i t h a/c CHOICE of CAR FLEET £ SIZE •average f l e e t weight of f a m i l y f + " ) (-1 c--> C-) G a s o l i n e p r i c e s • r e g i o n a l v a r i a b l e s n o n - l i n e a r non-1inear L i f e - c y c l e s t a g e •middle age c o u p l e CHOICE of NUMBER f_ of CARS OWNED p-— j r - T I n d i r e c t c o n s e r v a t i o n a c t i v i t i e s •keep r e c o r d of gas consumption Wealth •annual e a r n i n g s A v a i l a b i l i t y of p u b l i c t r a n s i t • c i t y s i z e Employment s t a t u s • w i f e ' s employment D r i v i n g needs •weekly commuting F a m i l y c o m p o s i t i o n •number i n f a m i l y F i g u r e 4.1 F a c t o r s I n f l u e n c i n g Automobile F l e e t C h o i c e . (Model d e r i v e d from e m p i r i c a l a n a l y s i s and c o n c e p t u a l development) * d i d not expect but found. > d i d expect and found. 1 1 5 c o n t i n g e n t upon c l i m a t i c f a c t o r s (coldness) as w e l l as consumer p e r c e p t i o n of p u b l i c t r a n s p o r t a t i o n e f f i c i e n c y . I t was s u r p r i s i n g to f i n d that consumers l i v i n g i n s i n g l e f a m i l y detached homes n e i t h e r own more ca r s nor own bigger c a r s . T r a v e l needs There i s a "feed-back" r e l a t i o n s h i p between the number of c a r s owned and the average car weight. T h i s i m p l i e s that consumers tend to d i v e r s i f y c a r - t y p e s . In other words, owners of m u l t i p l e c a r s , were l i k e l y to buy d i f f e r e n t s i z e s of c a r s . F u r t h e r , a i r - c o n d i t i o n e d c a r s and o l d e r c a r s are, g e n e r a l l y , bigger and h e a v i e r . Work r e l a t e d t r a v e l needs (weekly commuting, i f car used at work) p o s i t i v e l y a f f e c t the number of c a r s owned but do not a f f e c t average car weight. On the other hand, ownership of r e c r e a t i o n a l items (camper, boat et c . ) Generates a need f o r bigger v e h i c l e s but does not warrant more c a r s . T h i s i n d i c a t e s that work r e l a t e d t r a v e l needs and l e i s u r e needs are s a t i s f i e d through d i f f e r e n t s e t s of o p t i o n s . A t t i t u d i n a l f a c t o r s The marginal impacts of a t t i t u d e s and p a r t i c i p a t i o n items on the number of c a r s owned i s s m a l l . However, most of the v a r i a t i o n i n average car weight i s accounted f o r by these items. More s p e c i f i c a l l y , consumer p r e f e r e n c e s f o r automobile s i z e has a negative impact on the average ca r weight, as w e l l as on the number of c a r s owned. S i m i l a r l y , p a r t i c i p a t i o n i n g a s o l i n e c o n s e r v a t i o n a c t i v i t i e s and p e r c e p t i o n of p u b l i c t r a n s i t e f f i c i e n c y n e g a t i v e l y a f f e c t the average ca r weight but p o s i t i v e l y a f f e c t the number of c a r s owned. 1 16 To summarize, the conceptual development pr o v i d e d i n Chapter II as w e l l as the e m p i r i c a l r e s u l t s i n t h i s chapter suggest that consumer car c h o i c e s are contingent on demographic, s i t u a t i o n a l , t r a v e l needs, and a t t i t u d i n a l f a c t o r s . The flow diagram in F i g u r e 4.1 i n d i c a t e s that f a c t o r s a f f e c t i n g the number of c a r s owned may not n e c e s s a r i l y a f f e c t average car weight. I t i s a l s o noted that demographic and s i t u a t i o n a l f a c t o r s are b e t t e r p r e d i c t o r s of the number of c a r s owned, while a t t i t u d i n a l and p a r t i c i p a t i o n items b e t t e r p r e d i c t average car weight. 4.5 Pol i c y I m p l i c a t i o n s and Future Mode11ing Research. The r e s e a r c h reported i n t h i s study can not be c o n s i d e r e d a proper t o o l to analyze p u b l i c p o l i c i e s , however, i t i s p o s s i b l e to i d e n t i f y some p o l i c y i m p l i c a t i o n s from the m o d e l l i n g work. S i m i l a r l y , i t i s p o s s i b l e to i d e n t i f y s e v e r a l key areas f o r a d d i t i o n a l m o d e l l i n g of consumer automobile c h o i c e s and energy consumption. P u b l i c Pol i c y Issues In the short run (one or two y e a r s ) , p o l i c i e s based on g a s o l i n e p r i c e s intended to save g a s o l i n e may not be s u c c e s s f u l . T h i s i s p a r t i c u l a r l y e vident from comparative e f f e c t s of f a m i l y s i z e and income. Larger f a m i l i e s not only r e q u i r e a greater number of c a r s but a l s o c r e a t e a need f o r bigger c a r s . F u r t h e r , l o c a l , r e g i o n a l or even c u l t u r a l v a r i a t i o n s seem to bear a stronger r e l a t i o n s h i p to automobile 1 1 7 ownership than macro l e v e l g a s o l i n e p r i c e s . T h i s i m p l i e s , that i n c r e a s i n g g a s o l i n e p r i c e s may not r e s u l t in l a r g e g a s o l i n e s a v i n g s . T r a v e l needs and p o p u l a t i o n s i z e of urban c e n t r e s may have long term i m p l i c a t i o n s on g a s o l i n e consumption. I t may be concluded f a m i l i e s l i v i n g in l a r g e r urban c e n t r e s tend to own fewer and s m a l l e r c a r s than r u r a l f a m i l i e s because of the access to p u b l i c t r a n s p o r t a t i o n . T h i s i m p l i e s that p o l i c i e s which encourage the use of ef f i c i e n t and ef f e c t ive p u b l i c t r a n s p o r t a t i o n are l i k e l y to have m u l t i p l e g a s o l i n e c o n s e r v a t i o n b e n e f i t s and w i l l stand a b e t t e r chance of saving g a s o l i n e . On the other hand, l a r g e r commuting d i s t a n c e s to work r e s u l t i n g from urban sprawl may o v e r r i d e the b e n e f i t s of p u b l i c t r a n s p o r t a t i o n . Thus, p u b l i c p o l i c i e s that allow sprawl of urban development, may c o n f l i c t with p o l i c i e s that develop e f f i c i e n t p u b l i c t r a n s p o r t a t i o n systems. Future M o d e l l i n g E f f o r t s M o d e l l i n g research p r o v i d e s a s t r u c t u r e d approach to study consumer behaviour. T h i s study, however, was l i m i t e d to studying automobile ownership. There are s e v e r a l extensions p o s s i b l e . In Chapter I, i t was argued that g a s o l i n e consumption i s a f u n c t i o n of the number and s i z e of c a r s and car use v a r i a b l e s . T h i s study modelled number and s i z e of c a r s and i n d i c a t e d that g r e a t e r use r e s u l t s i n a g r e a t e r number of c a r s being owned. The f a c t o r s d i r e c t l y a f f e c t i n g use, however, were not i d e n t i f i e d . Thus, a car use model would provide more complete s t r u c t u r a l l i n k a g e s and help i n the 1 18 e s t i m a t i o n of g a s o l i n e consumption. One of the shortcomings of c r o s s - s e c t i o n a l consumer r e s e a r c h i s that i t ignores the dynamics of consumer c h o i c e s . L o n g i t u d i n a l consumer l e v e l data i s expensive to c o l l e c t and i t i s d i f f i c u l t to ensure response r e l i a b i l i t y . On the other hand, r e g i o n a l or n a t i o n a l aggregate data may not give accurate i n s i g h t s about micro consumer c h o i c e processes. I d e a l l y , car purchase (ownership and use) - new or o l d - or l o n g i t u d i n a l panel data may provide a s t a r t i n g p o i n t i n understanding consumer behaviour c o m p l e x i t i e s . There i s another form of dynamics i n v o l v e d i n car ownership d e c i s i o n s . If a car i s viewed as one of the many durables owned by consumers, then consumers must pursue t r a d e - o f f s between the ownership and use of these d i f f e r e n t d u rables. To date, n e i t h e r macro or micro economic models have been c o n s t r u c t e d to examine these c h o i c e t r a d e - o f f s . 1 19 References Akers, Fred C.(1968) Negro and White Automobile Buying Behavior: New Evidence J o u r n a l of Marketing Research, v o l . 5, pp. 283-290. B i r d w e l l , A l E.(1968) A Study of the I n f l u e n c e of Image Congruence on Consumer Choice J o u r n a l of Business, v o l . 41, pp. 76-88. Claxton John D. and McDougall Gordon G. H.0 983) The Household Automobile F l e e t : Dynamics of Ownership and usage. J o u r n a l of Consumer A f f a i r s ( i n p r e s s ) . Cragg, John G. and Uhler R. S.O 970) The Demand for Automobiles Canadian J o u r n a l of  Economics, v o l . 3, pp. 386-406. Degenais, M.G.(1975) A p p l i c a t i o n of a T h r e s h o l d Regression Model to Household Purchase of Automobile Review of Economic  and S t a t i s t i c s , v o l . 57, pp. 275-285. Dobson, Ric a r d o and Larson, Karen E.(1979) P s y c h o l o g i c a l and Socioeconomic C o r r e l a t e s of Automobile S i z e T r a n s p o r t a t ion Research Record, vol.723, pp. 7-11. Draper, N. R. and Smith, H. (1966) A p p l i e d Regression A n a l y s i s . Wiley, New York. Emmerson, P. and Downs, J . D.(1982) T r a v e l Change in Reading between 1962, 1971 and 1976 Transport and Road Research Laboratory, sup. Report 726. London, England. Evans, F r a n k l i n B.(1959) P s y c h o l o g i c a l and O b j e c t i v e F a c t o r s i n the P r e d i c t i o n of Brand Choice. J o u r n a l of Business, v o l . 32, pp. 340-369. Evans, F r a n k l i n B.0967) Ford versus C h e v r o l e t : Park F o r e s t R e v i s i t e d . J o u r n a l  of Business, v o l . 41, pp. 445-459. Friedman, M i l t o n ( l 9 5 7 ) A Theory of the Consumpt ion Funct ion P r i n c e t o n : P r i n c e t o n U n i v e r s i t y P r e s s . 1 20 Greene, David L. and Chen, E r i c C.K.(1981) Scappage and S u r v i v a l Rates of Passenger Cars and L i g h t Trucks i n the U.S., 1966-77 T r a n s p o r t a t i o n  Research, v o l . 15A, pp. 383-389. Greenlees, John S.(1980) G a s o l i n e P r i c e s and Purchases of New Automobiles Southern Economic J o u r n a l , v o l . 47, pp. 167-178. Grubb, Edward L. and Hupp, Gregg(l968) P e r c e p t i o n of S e l f , G e n e r a l i z e d Stereotypes and Brand S e l e c t i o n . J o u r n a l of Marketing Research, v o l . 5, pp. 58-63. Hensher, David A. and Johnston, L e s t e r W.(1981) A p p l i e d D i s c r e t e Choice M o d e l l i n g A H a l s t e d press book, N.Y.:John Wiley and sons. Hess, Alan C.(1977) A Comparison of Automobile Demand Equations Econometrica, v o l . 45, pp. 683-702. Hocking, R. R. (1976) The A n a l y s i s and S e l e c t i o n of V a r i a b l e s i n L i n e a r R e g r e s s i o n . B i o m e t r i c s , v o l . 32, pp. 1-49. I t o , Rikuma(l967) D i f f e r e n t i a l A t t i t u d e s of New Car Buyers, J o u r n a l of  A d v e r t i s i n g Research, v o l . 7, pp. 38-42. Johnston, T e r r y d 978) A C r o s s - s e c t i o n A n a l y s i s of New and Used Automobiles i n U.S. Economic I n q u i r y , v o l . 16, pp. 531-548. K a s s a r j i a n , H a r o l d H.(1971) P e r s o n a l i t y and Consumer Behavior: a Review, J o u r n a l  of Marketing Research, v o l . 8, pp. 409-418. Katz, R.O 980) Energy Conservation from a Consumer P e r s p e c t i v e : A S t a t e of the Art Review Working paper, I n s t i t u t Fur Marketing, W e s t f a l r s c h e E i l h e l m s - U n i v e r s i i t a t Munster, Fed. Rep. of Germany. Lave, C h a r l e s and T r a i n , Kenneth(1979) A Disaggregate model of Auto-type Choice T r a n s p o r t a t i o n Research, v o l . 13A, pp. 1-9. Lerman, S. R. and Ben-Akiva, M.(1976) Disaggregate B e h a v i o r a l Model of Automobile Ownership T r a n s p o r t a t ion Research Record, v o l . 569, pp. 34-55. 121 Mallows, C. L. . ( 1 973) Some Comments on Cp. Technometrics, v o l . 15, pp. 661-675. Manski, Charles and Sherman, Leonard(1980) An E m p i r i c a l A n a l y s i s of Household Choice among Motor V e h i c l e s T r a n s p o r t a t ion Research, v o l . 14A, pp.349-366. McDougall, Gordon H. G., Claxton, John D., R i t c h i e , Brent J . R. and Anderson, Denis C.(1981 ) Consumer Energy Research: A Review J o u r n a l of Consumer  Research, v o l . 8, pp. 343-354. McDougall, Gordon H. G., R i t c h i e , Brent J . R. and C l a x t o n , John D. (1980) Energy Consumption and Conservation  P a t t e r n s Summary Report. Consumer Research and E v a l u a t i o n Branch, Consumer and Corporate A f f a i r s , Canada, 1980. McFaden, D a n i e l ( l 9 7 8 ) M o d e l l i n g the Choice of R e s i d e n t i a l L o c a t i o n . i n , K a r l q v i s t , Anders (eds.) S p a t i a l I n t e r a c t i o n : Theory  and Planning Models, pp. 75-96. McFadden, D a n i e l ( l 9 7 6 ) A Comment on D i s c r i m i n a n t A n a l y s i s "versus" L o g i t A n a l y s i s Annals of Economic and S o c i a l Measurement, v o l . 5/4, pp. 511-523. McFadden, D a n i e l ( l 9 7 4 ) C o n d i t i o n a l L o g i t A n a l y s i s of Q u a l i t a t i v e Choice Behavior, i n Zarembka, P.(ed.) Front i e r s i n  Econometrics, N.Y.: Academic press, pp. 105-142. M i t c h e l s o n , Ronald L. and Gauthier, Howard(l979) Examination of Psy c h o p h y s i c a l F u n c t i o n i n T r a v e l Mode-choice Behavior T r a n s p o r t a t i o n Research Record, v o l . 750, pp. 27-33. M o d i g l i n i , Franco and Brumberg R. (1955) U t i l i t y A n a l y s i s and Aggregate Consumption F u n c t i o n : An I n t e r p r e a t a t i o n of C r o s s - s e c t i o n a l Data, i n K u r i h a r a , K. K. (ed.) Post  Keynesian Economics. London: George A l l e n & Unwin. Morrison, Donald F. (1976) M u l t i v a r i a t e S t a t i s t i c a l Methods Second ed. New York: McGraw H i l l co. 1 22 M u f t i , Rasin K. and Munson, Michael J.(1979) Approach to A s s e s s i n g the Impact of Energy Conservation P o l i c i e s on T r a n s p o r t a t i o n Demand T r a n s p o r t a t i o n Research Record, v o l . 726, pp. 1-7. Murphy, P a t r i c k E. And S t a p l e s , W i l l i a m A.(1979) A Modernized Family L i f e C y c l e . J o u r n a l of Consumer  Research, v o l . 6, pp.12-22. Recker, W. W. and Golob, T. F.( 197.6) An A t t i t u d i n a l Modal Choice Model T r a n s p o r t a t ion Research, v o l . 10, pp. 299-310. R i t c h i e , Brent J . R., McDougall, Gordon H. G. and C l a x t o n , John D.(1981 ) Co m p l e x i t i e s of Household Energy Consumption and Conservation J o u r n a l of Consumer Research, v o l . 8, pp. 233-242. Ruff i n , M a r i l y n D. and T i p p e t , K a t h e r i n e S.(1975) S e r v i c e - l i f e Expectancy of Household A p p l i a n c e s : New Estimates from the USDA Home Economics Research  J o u r n a l , v o l . 3, pp. 159-170. Singore, J e a n i e D. and K a s s a r j i a n , Harold H. (1981) The R e l a t i o n s h i p between O p i n i o n s , S o c i a l Values and Behaviours during 1979 G a s o l i n e C r i s i s : An E x p l o r a t o r y Attempt. In Bernhardt, Kenneth, D o l i c h I r a e t . a l . (eds.) The Changing Market ing Envi ronment: New  T h e o r i e s and A p p l i c a t i o n , pp. 295-298. Ster n , Paul C. and Gardner, G e r a l d T.(1981) P s y c h o l o g i c a l Research and Energy P o l i c y American  P s y c h o l o g i s t , v o l . 36, pp. 329-342. T a r d i f f , Timothy J.(1980) V e h i c l e Choice Models: Review of Previous S t u d i e s and D i r e c t i o n s f o r F u r t h e r Research T r a n s p o r t a t ion Research, v o l . 14A, pp. 327-335. A T e c h n i c a l Background Document f o r Automotive Fu e l Economy The T r a n s p o r t a t i o n Energy Management Program and O n t a r i o M i n i s t r y of Energy, O n t a r i o : Downsview, 1982. T r a i n , Kenneth(1980) A S t r u c t u r a l L o g i t model of Auto Ownership and Mode Choice Review of Economic S t u d i e s , v o l . 47, pp. 357-370. 1 23 T r a i n , Kenneth(1979) Consumers' Responses to F u e l - E f f i c i e n t V e h i c l e s : A C r i t i c a l Review of Econometric Stud i e s T r a n s p o r t a t i o n , v o l . 8, pp. 237-258. W a l l i n , R. J . and Wright, P. H.0974) F a c t o r s which I n f l u e n c e Modal Choice Traf f i c Q u a r t e r l y , v o l . 28, pp. 271-289. W e s t f a l l , Ralph(1971) P s y c h o l o g i c a l F a c t o r s i n P r e d i c t i n g Product Choice, J o u r n a l of Marketing, v o l . 26, pp. 34-40. Wiseman, Frederick(1971) A Segmentation A n a l y s i s on Automobile Buyers during the New Model Year T r a n s i t i o n P e r i o d . J o u r n a l of  Marketing, v o l . 35, pp. 42-49. White, Kenneth J . (1978) A General Computer Program for Econometric Methods SHAZAM. Econometrica, pp. 239-240. Wood, P. S.(1981 ) The E f f e c t of S p e c i f i c V e h i c l e Maintenance Measures on  F u e l Consumption The T r a n s p o r t a t i o n Energy Management Program and O n t a r i o M i n i s t r y of Energy, O n t a r i o : Downsview. 1 24 APPENDIX I U t i l i t y Maximization Model A b a s i c s t r u c t u r e of t h i s model i s d i s c u s s e d i n Hess(l977) and Cragg and Uhler(1970). The f o l l o w i n g a n a l y s i s p r o v i d e s some support to the d i s c u s s i o n i n S e c t i o n 2.2. Consider a u t i l i t y f u n c t i o n with a flow of consumption, C ( t ) , a stock of p h y s i c a l a s s e t s K ( t ) , and r e a l money balances M ( t ) / P ( t ) . The u t i l i t y f u n c t i o n to be maximized, a c c o r d i n g to l i f e - c y c l e model i s , T t (1) V(0) = L V [ C ( t ) , K ( t ) , M ( t ) / p ( t ) ] / ( 1 + T ) t = 0 where 7 i s a l i f e - t i m e d i s c o u n t r a t e , T i s l i f e - c y c l e i n t e r v a l and V(0) i s value of u t i l i t y f u n c t i o n at t=0. The stock K(t) can be broken i n t o three c a t e g o r i e s , automobile stock, A ( t ) , housing stock, H(t) and other durable stock, D ( t ) . Assume that each stock has 5a, 8h and 6d d e p r e c i a t i o n r a t e s . Thus, the investment i d e n t i t y would be K(t) = A(t) + H(t) + D(t) (2) I ( t ) = A(t) - (1 - 6 a ) A ( t - D + H(t) - (1 - 5h)H(t-1) + D(t) - (1 - 6 d ) D ( t - l ) where I ( t ) i s a gross investment i n year t . If p ( c , t ) , p ( a , t ) , p(h,t) and p(d,t) are p r i c e s a s s o c i a t e d with flow of consumption, automobile, housing and other d u r a b l e s , then l i f e t i m e budget c o n s t r a i n t can be represented by 1 25 T ( 3 ) Z { Y ( t ) - p ( c , t ) C ( t ) t t=0 - p ( j , t ) [ K ( t ) - ( 1 - 6 j ) K ( t - 1 ) ] - [M(t)-M(t-1)]}/(1+i) (j=a,d,h) where Y(t) i s nominal income in year t . I t i s necessary to assume V ( 0 ) to be twice d i f f e r e n t i a b l e and continuous w i t h i n the r e l e v a n t range. T h i s assumption, of course, can be met by c o n s i d e r i n g V ( 0 ) to be the lowest p o s s i b l e and then s c a l i n g the remaining goods by that u n i t . In a t h e o r e t i c a l d i s c u s s i o n , t h i s issue i s not a problem, however, in e m p i r i c a l r e s e a r c h t h i s assumption i s i m p r a c t i c a l . I t i s a l s o assumed that there are no r e s t r i c t i o n s on l e n d i n g or borrowing. Thus, maximizing V(0) s u b j e c t to ( 3 ) y i e l d s the f o l l o w i n g f i r s t order c o n d i t i o n s : (4) 3 c [ C ( t ) , K ( t ) , M ( t ) / p ( t ) ] / 6 - Xp(t)/0 = 0 (5) 9 a [ C ( t ) , A ( t ) , H ( t ) , D ( t ) , M ( t ) / p ( t ) ] / 6 - X [ p ( a , t ) - ( 1 - 5 a ) p ( a , t+ 1 ) / ( 1 + i ) ]/<p = 0 ( 6 ) 9 h [ C ( t ) , A ( t ) , H ( t ) , D ( t ) , M ( t ) / p ( t ) ] / 0 - X [ p ( h , t ) - ( 1 - 6 h ) p ( h , t + 1 ) / ( 1 + i ) ] / 0 = 0 ( 7 ) 9 d [ C ( t ) , A ( t ) , H ( t ) , D ( t ) , M ( t ) / p ( t ) ] / 8 - X [ p ( d , t ) - ( 1 - 6 d ) p ( d , t + 1 ) / ( l + i ) ] / 0 = 0 ( 8 ) 9 m / p [ C ( t ) , A ( t ) , H ( t ) , D ( t ) , M ( t ) / p ( t ) ] / d - Xp(t) i / ( 1+i )<}> = 0 t where # = ( 1 + 7 ) i s u t i l i t y f u n c t i o n d i s c o u n t f a c t o r , t 0=(l+i) i s a market d i s c o u n t r a t e , and 9 c , 9 a , 9 h , 9d and 9m/p represent the p a r t i a l d e r i v a t i v e s of the u t i l i t y f u n c t i o n with respect to flow of consumption, automobile, housing, other d u r a b l e s and r e a l money balances and X i s Lagrangian m u l t i p l i e r . 1 26 An important c o n c l u s i o n that can be reached from the above c o n d i t i o n s f o r e m p i r i c a l r e s e a r c h i s that the a p p r o p r i a t e p r i c e s f o r auto, housing and other durables are o p e r a t i n g and f i x e d c o s t s and not purchase p r i c e . However, i f l e n d i n g and borrowing c o n s t r a i n t s are imposed, the purchase p r i c e a l s o becomes b i n d i n g . For e q u i l i b r i u m , f o l l o w i n g c o n d i t i o n s must be s a t i s f i e d , 9 c [ C ( t ) , K ( t ) , M ( t ) / p ( t ) ] p ( t ) (9) 3 a t C ( t ) , A ( t ) , H ( t ) , D ( t ) , M ( t ) / p ( t ) ] p ( a , t ) - ( l - S a ) p ( a , t + l ) / ( 1 + i ) 9 m / p [ C ( t ) , K ( t ) , M ( t ) / p ( t ) ] X6 ip( t ) / ( 1 +i ) <t> Note that p(a,t) - (1-6a)p(a,t+1)/(1+i) i s a user - annual o p e r a t i n g and f i x e d - c o s t of an automobile f l e e t . To be c o n s i s t e n t with the u t i l i t y maximization, v a r i a b l e s i n these equations must be measured in the same p r i c e s . T h i s w i l l g ive a demand f u n c t i o n f o r auto (10) A(d,t) = <t>[u(a,t)/7r,u(d,t)/7T,u(h,t)/7r,u(m,t) ,Y( t ) / i r ] where 7r r e p r e s e n t s p ( t ) and u ( a , t ) , u ( d , t ) , u(h,t) and u(m,t) repr e s e n t o p e r a t i n g c o s t s a s s o c i a t e d with auto, durables, housing and r e a l money stocks r e s p e c t i v e l y . I t i s noted by Hess(l977) that " i n terms of e x t r a c t i n g p r o p e r t i e s of a s i n g l e demand equation which i s nested in a set of demand equations, e x p r e s s i o n (10) i s about as f a r as i t can be taken. I t p r o v i d e s two p i e c e s of i n f o r m a t i o n , v i z . the l i s t of arguments and the 1 27 a p p r o p r i a t e way to measure them." However, i f the automobile i s con s i d e r e d a d i s c r e t e u n i t , there are some important non-continuous e f f e c t s on the u t i l i t y f u n c t i o n . These e f f e c t s p r ovide f u r t h e r i m p l i c a t i o n s about e m p i r i c a l r e s e a r c h . 1 28 APPENDIX II E s t i m a t i o n of multinomial l o g i t parameters with binomial l o g i t computer program. At the present time SHAZAM(White, 1978) i s only e a s i l y a c c e s s i b l e computer program a v a i l a b l e to estimate maximum l i k e l i h o o d parameters of the l o g i t model at UBC. SHAZAM, however, can only provide estimates f o r binomial forms of the l o g i t and not multinomial model. In t h i s appendix, a t h e o r e t i c a l j u s t i f i c a t i o n and o p e r a t i o n a l methods are p r o v i d e d f o r e s t i m a t i o n procedure. T h e o r e t i c a l c o n s i d e r a t i o n McFadden(1978) has g e n e r a l i z e d the cumulative d i s t r i b u t i o n f u n c t i o n f o r independent e r r o r s i n d i s c r e t e c h o i c e s to (1) F ( e 1 f ek+]) = exp[-G(exp-e ,...,exp-e k )] where G i s s p e c i f i e d to be a non-negative homogeneous f u n c t i o n of degree 1 , and tends towards +°° when any of i t s arguments tends towards +°°, and k+1 are a l t e r n a t i v e c h o i c e s e t . The marginal d i s t r i b u t i o n with respect to each i s a u n i v a r i a t e extreme-value d i s t r i b u t i o n , (2) F j ^ e j ^ = e x P ( " c j e x P ~ e j ) where C j = G(3. . , . . . , 3 . } ) with 3. . = 1 i f i = j and 3. . = 0 otherwise. If the ^th household maximizes i t s u t i l i t y by choosing j t h a l t e r n a t i v e , McFadden(1978) has shown that the necessary c o n d i t i o n f o r maximization i s 1 29 exp(V^ )G.j [exp(V 1 , . . . ,exp(V k + 1 ) ]  p j = G[exp(V ),...,exp(V k + )] where p^ i s p r o b a b i l i t y a s s o c i a t e d with j t h a l t e r n a t i v e , G^ i s p a r t i a l d e r i v a t i v e of G with respect to j t h argument and V_j i s u t i l i t y a s s o c i a t e d with ^th a l t e r n a t i v e . If one assumes a simple a d d i t i v e s t r u c t u r e f o r G, i . e . G [ e x p ( V 1 , . . . , e x p ( V k + 1 ) ] = exp(V 1) + ... + e x p ( V k + 1 ) . Then, note that G_j = 1 f o r a l l j , s i n c e p a r t i a l d e r i v a t i v e s are with respect to j t h argument. Thus, P i j ' t n e p r o b a b i l i t y that the ^th f a m i l y makes j t h choice i s (3) p.. = exp(V. . ) / 4 1 e x p ( V . , ) . 1 J -1- J 1 = 1 I t i s next assumed that the u t i l i t y (V. .) d e r i v e d from i ] choosing the j t h a l t e r n a t i v e i s (4) v.. = t?Z . + 0.X. + e . . 1 D D D I 13 where Zj i s an n-element v e c t o r of a l t e r n a t i v e s p e c i f i c a t t r i b u t e s , X^ i s an m-element v e c t o r of consumer c h a r a c t e r i s t i c s r e l a t i n g to the ^th o b s e r v a t i o n , 6 and 0j are m-element and n-element v e c t o r s of unknown parameters. T h i s p a r t i c u l a r form of u t i l i t y f u n c t i o n was chosen f o r s t a t i s t i c a l e s t i m a t i o n purpose, to accomodate consumer c h a r a c t e r i s t i c s i n the demand f u n c t i o n . These, c h a r a c t e r i s t i c s i n t urn may be c o n s i d e r e d the set of c o n s t r a i n t to the consumer u t i l i t y maximization problem and s p e c i f i c c o e f f i c i e n t s r e f l e c t 1 30 Lagrange m u l t i p l i e r e f f e c t f o r each c h o i c e . Since the a l t e r n a t i v e s p e c i f i c a t t r i b u t e i n f o r m a t i o n (e.g. o p e r a t i n g cost per c a r) was not a v a i l a b l e i n t h i s study, i t was assumed Zj to be constant across c h o i c e s and i t can be d e l e t e d from (4). In other words, constant term i n above model r e f l e c t s r e l a t i v e consumer u t i l i t y with respect to s p e c i f i c consumer c h o i c e s . S u b s t i t u t i n g (4) i n (3), the f o l l o w i n g e x p r e s s i o n can be obt a i n e d . (5) p.. = exp(j3.X. + e . . ) / k£exp(0,X. + e . , ) . * "] l i ] i=i 1 1 3 - l It now remains to be shown that (5) can be estimated using a computer package SHAZAM that only handles b i n a r y v a r i a b l e s . There are, however, two p o s s i b l e approaches f o r using SHAZAM to o b t a i n multinomial l o g i t parameter estimates. Each of these approaches and t h e i r m e r i t s w i l l be d i s c u s s e d below. Approach 1 For k+1 c a t e g o r i e s , c o n s t r u c t k binomial (0,1) random dependent v a r i a b l e s (e.g. y 1 , ••• such that Then, one may use the SHAZAM program to o b t a i n Maximum l i k e l i h o o d parameters of the k dependent v a r i a b l e s . In t h i s f o r m u l a t i o n , while the c o v a r i a n c e s between the dependent v a r i a b l e s are c o n s t r a i n e d to be zero the co v a r i a n c e s between the l a t e n t v a r i a b l e s are not so r e s t r i c t e d . More s p e c i f i c a l l y , these c o v a r i a n c e s are E [ P i / p k + 1 , P j / p k + l ] * 0 T h i s , i n t u r n , w i l l i n t r o d u c e b i a s i n t o the parameter est i m a t e s . 131 T h i s b i a s , of course, can be minimized by e s t i m a t i n g the k seeming u n r e l a t e d equations simultaneously, t h i s p o s s i b i l i t y , however, appeared not to be f e a s i b l e f o r the scope of t h i s study. Approach 2 For k+1 c a t e g o r i e s , c o n s t r u c t k, binomial (0,1) random dependent v a r i a b l e s such that y^ = 1 i f category code i s 1, 0 i f category code i s (k+1), (say n 1 cases e i t h e r have code 1 or code k+1 ) , y^ = 1 i f category code i s k, 0 i f category code i s (k+1), (say n^ cases e i t h e r have code k or code k+1 ) . In matrix n o t a t i o n t h i s scheme would be (6) F(y) = l o g ( p x / p k + 1 ) = X0 + e where y i s 1x(n l+n2+... vect o r of observed Y i ' s , p x i s l a t e n t p r o b a b i l i t i e s of (x=1,2,... ,k), P k + 1 i s the l a t e n t p r o b a b i l i t i e s of the k+1 c h o i c e , X i s (n 1+n 2+... +n k )x (m1+m2+. . . + m j c ^ matrix of independent v a r i a b l e s with m1, m2, ...m^  r e s p e c t i v e p r e d i c a t o r s e t s along t diagonal and zero a l l other p l a c e s . 1 32 P i c t o r i a l l y , the X matrix i s X1 . 0 0 X k 0 i s 1x(ml+m2+... + m k ) matrix of unknowns to be estimated, e i s 1x(n 1+n 2+... + n k) matrix of e r r o r values d i s t r i b u t e d a c c o r d i n g to the W e i b u l l d i s t r i b u t i o n , X 1 f...X k are sub-matrices of X matrix. It can be shown that (6), the binomial l o g i t e s t i m a t i o n problem i s e q u i v a l e n t to the g e n e r a l i z e d extreme value problem such as (5) . Note that f o r n 1 cases, l o g ( p 1 / p k + 1 ) = X ^ + e 1 (7) n R cases l o g ( p k / p k + 1 ) = X R/3 k + * k 1 33 The system of equations i n (7) can be w r i t t e n as P/Pk-H = exp(X 1/3 1 + e,) P 2 / P k + 1 = e x p ( X 2 0 2 + e 2 ) (8) P k / P k + 1 = exp(X k/3 k + e k) where &^ , 0 2,... 0 k are segments' of |3, the v e c t o r of estimated c o e f f i c e n t s such that 0 i s 1 xm , ' j32 i s 1xm2, ... /3k i s 1 xm^. Fur t h e r , the sum of a l l the equations i n (8) w i l l r e s u l t i n (9) ( P l+p 2+. . . + p k ) / p k + 1 = k£4xp(X./J. + e.) . However, Pi + P 2 + ' " + Pk + Pk +1 = 1 or Pi + P 2 + - " + Pk = 1 " P k +1 S u b s t i t u t i n g above i n t o (9), r e s u l t s i n (1 " P k + 1 ) / P k + 1 " * & * P < V i + e . ) . S o l v i n g f o r P k + 1 r e s u l t s i n , (10) p k + 1 = 1/[1 +kVexp(Xi(3i + e . ) ] . i = 1 Fu r t h e r , to accomplish e s t i m a t i o n , the f o l l o w i n g r e s t r i c t i o n i s imposed ^k +1 = 0 T h i s i m p l i e s t h a t , e x p ( X k + 1 / 3 k + 1 ) = 1 Thus, (10) i s a g e n e r a l i z e d extreme value f u n c t i o n with the f o l l o w i n g form: 1 34 p. = exp(X./3.) / k£exp(X,/3. + e .) , j = 1 J J J i=1,...,k+1, subj e c t to <Vl = °-Using t h i s approach, the parameters i n Table 4.1 and Table 4.4 are estimated. A l l these approaches, however, assume that v a r i y ^ = v a r ( y 2 ) = ... = v a r ( y k ) . In s t a t i s t i c a l terms, t h i s assumption i s not r e s t r i c t i v e , s i n c e a l l y^ are (0,1) v a r i a b l e s . In e m p i r i c a l r e s e a r c h , however, t h i s assumption needs to be t e s t e d p r i o r to using the e s t i m a t i o n procedure. In t h i s study, y 1 was one, i f the f a m i l y owned one car and zero i f the f a m i l y owned two c a r s and s i m i l a r l y , y 2 was one, i f the f a m i l y owned three c a r s and zero i f the f a m i l y owned two c a r s . It was observed that v a r ( y ) = 0.265 v a r ( y 2 ) = 0.534 In a b s o l u t e terms, v a r ( y 1 ) * v a r ( y 2 ) but i t i s not p o s s i b l e to t e s t hypothesis concerning v a r ( y 1 ) and v a r ( y 2 ) s i n c e d i s t r i b u t i o n of y 1 and y 2 are b i n o m i a l . An approximation, based on c e n t r a l l i m i t theorem, would be to assume v a r ( y 1 ) and v a r ( y 2 ) random v a r i a b l e s with x 2 d i s t r i b u t i o n . The n u l l h ypothesis of v a r ( y 1 ) = v a r ( y 2 ) i s r e j e c t e d ( F o ^=2.02). Since the v a r i a n c e f o r two vs. three car owning f a m i l i e s i s higher than the the v a r i a n c e f o r the two vs. one c a r s , estimated parameters of the former must be i n t e r p r e t e d with c a u t i o n . 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0096024/manifest

Comment

Related Items