Household Expenditure on Food Away from Home by Type of Meal in Malaysia

Transcription

1 Pertanka J. Soc. Sc. & Hum. 18 (2): (2010) ISSN: Unverst Putra Malaysa Press Household Expendture on Food Away from Home by Type of Meal n Malaysa Gaz Nurul Islam 1*, Ta Shzee Yew 2, Nk Mustapha R. Abdullah 2 and Kusar Mohd Noh 1 1 Insttute of Agrcultural and Food Polcy Studes, Putra Infoport IOI Resort, Jalan Kajang-Puchong, Malaysa 2 Faculty of Economcs and Management, Unverst Putra Malaysa, UPM, Serdang, Selangor, Malaysa * E-mal: gaznurul236@gmal.com ABSTRACT Ths paper analyses the household food-away-from-home (FAFH) expendture pattern n Malaysa. For ths purpose, the Tobt model was used to quantfy the responsveness of households expendture on FAFH to changes n ther ncome and the household characterstcs. The results show that households FAFH expendture has ncreased due to rse n ncome and changng lfestyle. Meanwhle, the household ncome has statstcally sgnfcant nfluence on the FAFH expendture for all types of meals, except for breakfast, although the FAFH expendture for breakfast s postve. The postve effect of the total household ncome shows Malaysan households spent more by eatng out as they have to spend more tme at work and less tme to prepare food at home. Breakfasts and lunches away from home have postve and nelastc ncome. The results ndcate that household members have less choce but to consume breakfast and lunch away from home as ther workplaces are usually far away from ther homes and thus consume these meals at home are not cost-effectve. They have greater flexblty n makng decsons whether to consume at home or away from home for dnner and other meal. The estmated condtonal and uncondtonal ncome elastcty for the households FAFH expendture for all types of meals shows that the FAFH expendtures by Malaysan households are ncome nelastc. Ths mples that the growth n the FAFH sector wll largely be drven by household demographcs, ethnc characterstcs and regon n Malaysa. The government should take approprate measures to ensure that the meals are of hgh nutrtous values, safe, and reasonably prced. Keywords: Condtonal and uncondtonal elastcty, food away from home, Malaysa, type of meal INTRODUCTION Eatng away from home for Malaysan households has ncreased over the years, and ths has drawn a sgnfcant nterest by polcy makers for many reasons. Ths country has acheved hgh ncome growth and experenced rapd structural transformaton and urbanzaton n the recent years. The changes n socoeconomc and demographc structure have also occurred the average household sze has been fallng (from 5.2 n 1980 to 4.3 persons n 2005), whle the percentage of the populaton n the 65 years and above age category s ncreasng (Department of Statstcs Malaysa, 2006). Women workng outsde home also showed an ncreasng trend from 44.5% to 46.7% between 2000 and 2005 (Mnstry of Fnance, 2004). Receved: 6 August 2009 Accepted: 22 January 2010 * Correspondng Author

2 Gaz Nurul Islam et al. Malaysa s a mult-ethnc country consstng of three domnant ethnc groups, namely the Malays (65%), Chnese (26%), and Indans (8%). Each ethnc group has ther own food habts and the choce of the consumers may vary wdely among these heterogeneous ethnc groups. In Malaysa, the expendture on food at home (FAH) was found to have declned from a share of 36.2% of the total household food budget n 1973 to 20.1% n 2004, whle spendng on food away from home (FAFH) rose from 8.9% n 1973 to 10.5% n These fndngs ndcate that Malaysans have changed ther preference for food away from home at the expense of havng meals at home. The market worth of consumer food servce transactons was RM16,312 mllon (US$4,315 mllon n the current value n 2003) whch grew from 22% to 39% from1999 to 2003 (Lee and Tan, 2007). The ncreasng trend of home delvery and take-away food sector clearly demonstrates that the Malaysan households are spendng less tme n preparng FAH, a reflecton of busy work schedules outsde the home. There s a rapd growth of food servce facltes, such as fast food restaurants and street food stalls, avalable throughout the country. At the same tme, a varety of food tems and the servce facltes are now avalable n the FAFH sector. Ths wll lkely to have contnuous mpacts on the dstrbuton, marketng, food servce system, and the nutrtonal ntake. Wth growng urbanzaton, t s expected that there wll be sgnfcant changes n future food consumpton n terms of detary habts and food preferences n Malaysa. The mportant concerns for FAFH are related to the nutrtonal value, food safety and the ambent envronment of the eateres. The lterature revewed reveals that very lttle research has been conducted to examne the FAFH consumpton pattern n ths regon. Most studes on the FAFH expendture have dentfed ncome, race, household sze, and resdence of the households (Hanna and Carter, 1986; Lee and Brown, 1986; McCracken and Brandt, 1987; Soberon-Ferrer and Dards, 1991; Yen, 1993), seasonal effects, age, and status of parents (Hemstra and Km, 1995; Yen, 1993; Nayga and Capps, 1992) as mportant determnants of the FAFH expendture. Meanwhle, Res, Klne and Weaver (1987) found that the nutrent densty of FAFH was lower than FAH. McCracken and Brandt (1990) asserted that both the soco-economc and demographc varables have dfferental mpacts on the FAFH expendture facltes (dfferent types of eatng places) and have mportant mplcatons n analyzng household food demand. Reynolds and Goddard (1993) examned the determnants of FAFH based on the type of facltes and the types of meal n Canada. They found that the household choces on FAFH dffer by the type of meal. In Malaysa, studes on FAFH have mostly been done at the total expendture levels. Three studes (Ishda et al., 2003; Lee and Tan, 2007; Alas et al., 2006) looked at the FAFH n Malaysa usng the Household Expendture Survey data. Among other, Ishda et al. (2003) estmated the Engel s demand elastcty whle Lee and Tan (2007) and Alas et al. (2006) found that gender (male), age of household head, household sze and educaton are the sgnfcant varables. Meanwhle, ndvduals from urban areas appeared to have hgh FAFH demand as compared to those from the rural areas. Ths paper nvestgates the household expendture on the FAFH based on the type of meals (breakfast, lunch, dnner, and other meal) for the households n the urban and rural areas of Malaysa. The objectves of the study were to: (a) determne the consumpton trends and spendng patterns of Malaysan households on food away from home, and (b) estmate the responsveness of FAFH meals to changes n ncome and selected household demographc characterstcs. Ths paper s organzed as follows; Secton 2 descrbes the Model used n the study; the data and ther sources are descrbed n Secton 3; the results are dscussed n Secton 4; and the conclusons are presented n the fnal secton. 286 Pertanka J. Soc. Sc. & Hum. Vol. 18 (2) 2010

3 Household Expendture on Food Away from Home by Type of Meal n Malaysa MODEL The household producton theory (Becker, 1965; Lancaster, 1971) has been used n many studes as the underlyng theoretcal bass for analyzng the demand for food away from home (Stewart et al., 2004). The theory posts that households are maxmzng utlty n the consumpton of home-produced goods subject to a household producton functon, tme, and ncome constrants. The producton theory manly accounts for tme constrants n the household s decson makng process, and thus, the value of the homemaker s tme s an mportant determnant of FAFH expendture (Prochaska and Schrmper, 1973). As the opportunty cost of tme rses, households can be expected to purchase food tems that requre less tme to process at home. In ths case, food bought outsde home and not cooked at home would be consdered as FAFH consumpton. At the same tme, the respondents may report zero FAFH expendture. The estmaton of coeffcent wll be nconsstent when only observed postve purchase data are used to estmate consumpton behavour by the OLS regresson. The censored Tobt model s approprate because t allows the presence of zero observatons attrbutable to corner solutons. Zero shares are censored by an unobservable latent varable. The Tobt model s used to handle the censored sample,.e. a sample n whch nformaton on the dependent varable s avalable only for some observatons. Ths model s often used to nvestgate the determnants of household expendture (Soberon-Ferrer and Dards, 1991). In order to purchase an tem, t s assumed that households make two decsons. These decsons are, frst, whether to spend - a partcpaton effect and, second, how much to spend a quantty effect. The researchers used the standard Tobt model to specfy the estmatng equaton for the demand model for the FAFH consumpton expendture. In ther work, Deaton and Irsh (1984) assumed a lnear equaton, whle the Tobt formulas for both latent demand and observed demand are: Y * = β X + ε, (latent demand) [1] Y ** = max(y *, 0) (observed demand) [2] The maxmum lkelhood method was used to estmate the b coeffcents. Ths method assures large sample consstency, asymptotc normalty of the estmated coeffcents and conventonal tests of sgnfcance. The log lkelhood functon for the Tobt model s expressed as: Log L = Σ 0 log Ф ( ' b x /σ) + Σ + {log { [(y - ' b x)/ σ] log σ } [3] Where Ф and { are the cumulatve probablty dstrbuton functon, the densty functon of the standard normal varable and σ s the standard devaton of μ (Greene, 2003). The subscrpts 0 and + mean that summaton s performed over the sub-sample n whch dependent varable s zero and postve. The margnal effects of the Tobt model, wth censorng at zero and normally dstrbuted dsturbances, can be wrtten as follows (Greene, 2003:766): 2E5 y x? b U 2x ' b x c m = [4] Where b s the estmated coeffcent of the ' b x Tobt model correspondng to. x U c m s the v cumulatve dstrbuton functon and v s the standard devaton of the dstrbuton. A useful decomposton of 2 E ( y / x ) / 2 x was suggested by McDonald and Moftt (1980): 2E5 y x? = b # U ( 1 - m ( a + m )) + { ( a + m ) 2x where ' U = U ( b x / v ) = Φ a v [5] ] g and m = { / U Takng the two parts separately, ths leads to the decomposton of the slope vector nto: 2E5 y x? = Prob (( y 0) 2x 2 5 2E y x, y 2x? E y x, y 5 2 Pr ob y 2 0? 2 0? [6] 2x Pertanka J. Soc. Sc. & Hum. Vol. 18 (2)

4 Gaz Nurul Islam et al. Thus, a change n x has two effects, namely, ts effects on the condtonal mean of y * n the postve part of the dstrbuton, and ts effects on the probablty that the observaton wll fall n that part of the dstrbuton. From the estmated margnal effect, the condtonal elastcty of y wth respect to any x such as ncome, can be computed as follows: c Condtonal Elastcty f = 2E6y x@ E ( x ) 2x E6y x@ [7] The uncondtonal elastcty of y, wth respect to any x can be computed as: u Uncondtonal Elastcty f = 2E ( y ) E ( x ) [8] 2x Ey DATA SOURCE A combnaton of the stratfed and cluster samplng desgn was adopted. A detaled lst of household census and locatons (Enumeraton Blocks and lvng quarters) throughout Malaysa was obtaned from the Department of Statstcs Malaysa. At the same tme, the stratfcaton by states was done at the frst stage and nne states were randomly selected from the lst. Each selected state was further dvded nto urban and rural clusters whch were randomly selected. Wthn each selected cluster, Lvng Quarters (households) were randomly selected. The selected sample ncluded 31 Enumeraton Blocks n 9 states, wth a total sample sze of 283 Lvng Quarters consstng of 176 households from the urban areas and 107 households from the rural (Table 1). The survey used structured questonnare at the household level. The questonnare was dvded nto sectons for ts convenence to record household soco-demographc characterstcs and to record food expendture away from home. The expendture of all the members of households on FAFH was recorded by types of meals. The household members n Malaysa consume three man meals a day. In addton, lght foods such as snacks, tea, bscuts, etc. nbetween meals are also beng consumed. The expendtures on meals away from home made by household members were recorded at the end of the each day durng the survey week. In many of the sampled households, record keepng work was performed by housewves, but n some, ths task was done by the school or college-gong members of the households. The sampled households kept records on ther food expendtures for 7 consecutve days, and the duraton of survey was spread over one month to overcome logstcal problems of havng to cover all the respondents n the same week. The survey perod lasted from May 15 to June 15, State TABLE 1 The sampled households by urban and rural areas for FAFH survey Urban Rural All No. Percent No. Percent All Percent WP Selangor Kedah P. Pnang Sabah Sarawak Terengganu Johor N. Semblan All Pertanka J. Soc. Sc. & Hum. Vol. 18 (2) 2010

5 Household Expendture on Food Away from Home by Type of Meal n Malaysa The sample statstcs of the FAFH expendture accordng to the type of meal are presented n Table 2. Of the total sample, 241 households (82%) reported expendture on breakfast, 224 households (79%) on lunch, 197 households (70%) on dnner, and 118 households (42%) on other meals away from home durng the one-week perod (Table 2). Average weekly FAFH expendture by households on breakfast, lunch, dnner, and other lght meal were RM34, RM 47, RM43 and RM11, respectvely (Table 2). Meanwhle, the weekly average FAFH expendtures for consumng households were RM40, RM60, RM61, and RM26 for breakfast, lunch, dnner and other meal, respectvely. The expendture also vared between the urban and rural areas, whle the overall expendture for all types of meals n urban areas was hgher than that of the rural areas (Table 3). Ths study used household expendture as proxy for the demand of FAFH by the households. The household expendtures on breakfast, lunch, dnner, and other meals away from home durng a one-week perod were used as the dependent varables. The mportant explanatory varables are household ncome, whch s an average monthly ncome of earnng members of the households. Other explanatory varables used nclude educaton, household sze, ethnc, and regon (urban and rural). The household ncome was hypotheszed to have a postve effect on the FAFH expendture. In other words, as ncome ncreases, households may allocate more of ther ncome to FAFH expendture. Several soco-demographc varables were also ncluded n the estmated equatons to capture the taste and preferences of the household FAFH consumpton. Meanwhle, household sze (measured as the number of persons resdng n the household durng the survey week) was expected to have a negatve effect on FAFH expendture. The effects of age of household head (n years), educaton attanment of household head (n years) and the number of chldren below 12 years were ambguous. The studes by Capps Jr. and Park (1997) as well as McCracken and Brandt TABLE 2 Sample statstcs of FAFH by meal type Full sample Consumng households Meal type Mean Std. dev. Mean Std. dev. N a Breakfast Lunch Dnner Others a Sze of consumng households TABLE 3 Sample statstcs of FAFH by meal type by regon Urban Rural Meal type Mean Std. dev. N Mean Std.dev. N Breakfast Lunch Dnner Others Pertanka J. Soc. Sc. & Hum. Vol. 18 (2)

6 Gaz Nurul Islam et al. (1987) used the varables for ethncty to capture cultural and ethnc dfferences that nfluence purchase decsons. They found that there were dfferences between younger and older people n tastes, eatng habts, and lfestyles. Several dummy varables were also used n an attempt to capture the regonal (urban or rural) and racal (Malay, Chnese, Indan, or East Malaysan Bumputra) dfferences n the FAFH expendture by households n ths study. The sample statstcs of all the explanatory varables are presented n Table 4 below. RESULTS The standard Tobt model for each meal and for the overall FAFH expendture was estmated by the maxmum lkelhood technque. Fve equatons were ndependently estmated n one for each type of meal, namely breakfast, lunch, dnner and others, and another for the overall FAFH expendture. The dependent varable for each of these equatons conssts of per week expendture (RM) on FAFH by household members. Table 5 presents the results of the estmated equatons. The total household ncome s statstcally sgnfcant and has postve effect for the total FAFH expendture and the FAFH expendture on all types of meals except breakfast. The range of ncrease n the FAFH expendture s between RM0.02 to RM0.04 for every RM ncrease n the ncome per week. The postve effect of total household ncome shows that as ncome rses, households spend more by eatng out as expected probably because they have to spend more tme at work and less tme to prepare food at home. The results show that a few socodemographc varables were found to be statstcally sgnfcant n the overall FAFH TABLE 4 Sample statstcs of explanatory varables Varables Mean Std. dev. Mn Max Contnuous Total FAFH Expendture (RM/per week) FAFH Expendture on Breakfast (RM) FAFH Expendture on Lunch (RM) FAFH Expendture on Dnner (RM) FAFH Expendture on Others (RM) Total Food At Home (FAH) Expendture (RM) Household Income (RM) Household sze (No.) Household head Age (year) Educaton of household head (year) No. of Chldren (below 12 years) Dummy (yes=1, no=0) Urban households Malay household Chnese household Indan household Pertanka J. Soc. Sc. & Hum. Vol. 18 (2) 2010

7 Household Expendture on Food Away from Home by Type of Meal n Malaysa expendture, breakfast and lunch equatons and more of these varables are statstcally sgnfcant n the FAFH expendture equatons for dnner and other meals. The household members work places tend to be away from home. As a result, they prefer eatng breakfast and lunch at or close to ther workng places rather than at home n order to save tme and costs. However, for dnner and other meals, they may not be constraned by havng to return to ther work place after meal and thus, they have more flexblty n makng decsons whether to consume these meals at home or away from home. The effects of the household ncome (per week) on FAFH expendture on dnner and other meals are statstcally sgnfcant. For every RM ncrease n household ncome, the FAFH expendture on dnner and other meal ncreases by RM0.03 and RM0.01 per week. The FAH expendture has sgnfcant effect on the total FAFH expendture and FAFH expendture on all types of meals except lunch. The effects of the FAH expendture on the FAFH expendture are postve n all the equatons. For every RM ncrease n the FAH expendture, the FAFH expendture ncrease between RM0.03 and RM0.2 per week. Smlarly, the ncrease n the TABLE 5 Maxmum lkelhood estmates of Tobt model: Food away from home expendtures by meal type Varables Total FAFH expendture Breakfast Lunch Dnner Other meal Constant (77.569) (23.397) ( ) ( ) (28.962) Household Income (RM/ month) ** (0.0194) ( ) *** ( ) ** ( ) ** (0.0058) Age of household head (Year) (0.0501) ( ) ( ) ( ) *** (0.0075) Household sze (number of household member) (5.5847) ( ) (2.2714) (2.1198) ** (1.8753) Number of chldren (below 12 years) (7.4815) (2.2814) (3.297) (3.7728) *** ( ) Educaton of household head (level of educaton) (2.6064) (0.7586) ( ) (1.3553) ( ) Expendture at home (RM/per month) *** ( ) ** (0.0174) ( ) ** (0.0301) *** ( ) Ethnc (dummy) Malay ( ) (16.764) ( ) ( ) (27.253) Chnese (53.307) * (16.606) ( ) * ( ) ( ) Indan (50.713) ( ) ( ) ** ( ) ( ) Regon (18.67) (5.007) (7.8126) * (8.3948) ** (7.1757) Standard errors n parentheses. * (**) (***) Sgnfcant at the 0.10, 0.05 and 0.01 level. Pertanka J. Soc. Sc. & Hum. Vol. 18 (2)

8 Gaz Nurul Islam et al. FAH expendture by households wll ncrease FAFH expendture smultaneously, and ths s probably because of the households preference changes for better qualty and more nutrtous food that cost more. Age of the household head was found to have sgnfcant and negatve effect only on the FAFH expendture for other meals. In other words, as the age of household head ncreases by one year, the households spent less eatng out for other meals, but the reducton s very small,.e. at RM0.02 per week. The household sze also has a sgnfcant and negatve effect only on the FAFH expendture for other meals. Once agan, as household sze ncreases, the tendency for households to spend more on man meals such as breakfast, lunch, and dnner but they spend about RM2 less eatng out for other meal. Greater number of chldren below 12 years n the households has sgnfcant and postve effect on FAFH expendture for other meal. For an addtonal chld, households spent more than RM7 per week on other meals, and ths s probably due to the fact that school gong chldren spend more on food at school canteens. Ths ncreases ther tendency to eat out between meals at fast food restaurants. The regonal dummy varable has sgnfcant and postve effects on the FAFH expendture on dnner and other meals. These results ndcate that the urban households spent on average RM15 per week on dnner and RM15 on other meals away from home. For the racal dummes, Chnese households spent on average RM28 per week on breakfast and RM38 per week on dnner away from home more than the Bumputra households n East Malaysa. Meanwhle, the Indan households spent sgnfcantly more (RM46 per week on dnner away from home) as compared to East Malaysan Bumputra households. The elastcty of the FAFH expendture was computed based on the Tobt equatons estmates. Table 6 presents the condtonal and uncondtonal ncome elastctes. The estmates for all types of meals are all less than one, except for the uncondtonal elastcty of other meals (2.042) whch ndcates that the FAFH expendture by Malaysan households are ncome nelastc. The mplcaton s that a one percent ncrease n ncome wll ncrease the FAFH expendture by less than one percent. Meanwhle, the condtonal ncome elastcty for the overall FAFH expendture, breakfast, lunch, dnner, and other meals are 0.22, 0.09, 0.31, 0.29 and 0.51, respectvely. As for the uncondtonal ncome elastcty, ther values for all types of meals are hgher than the correspondng condtonal values. The uncondtonal ncome elastcty ranged from 0.11 for breakfast to 2.04 for other meals. These values are consstent wth those estmated n other studes. For nstance, Byrne et al. (1996, p. 615) estmated the ncome elastcty values between 0.11 and 0.36, whle Ma et al. (2006) estmated the elastcty of meat for FAFH n Chna to be CONCLUSIONS In ths study, the household FAFH expendture n Malaysa was analyzed usng the econometrc model that accommodates zero observatons n the sample. The results of the study show that there are potentals for the FAFH ndustry to expand and prosper n Malaysa. As the Malaysan economy contnues to grow n the long run and n turns causng household ncome to rse, the household FAFH expendture wll also be ncreased as ndcated by the postve, albet nelastc expendture wth respect to ncome. In other words, ncome ncreases mply that household members need to spend more tme at work and less tme to prepare food at home. Urbanzaton wll also rase household expendture on FAFH. In partcular, the urban households were found to spend more on FAFH than the rural households, partcularly on dnners and other meals. There are dfferences n the FAFH expendture among the households of varous ethnctes. Meanwhle, the Chnese and Indan households expendtures on FAFH are relatvely hgher as compared to the Malay and East Malaysan Bumputra households. Although FAFH provders need 292 Pertanka J. Soc. Sc. & Hum. Vol. 18 (2) 2010

9 Household Expendture on Food Away from Home by Type of Meal n Malaysa TABLE 6 Condtonal and uncondtonal ncome elastcty of demand for FAFH Meal type Condtonal elastctya Uncondtonal elastcty Total elastcty Breakfast Lunch Dnner Others All meals a The condtonal elastcty or quantty elastcty s the elastcty of expected consumpton of consumng ndvdual wth respect to the ndependent varable (FAFH expendture). to take cognzance of the ethnc dfferences and cater ther servces accordngly, t wll be more sensble and approprate to provde FAFH that s acceptable to all the ethnc and relgous groups. At the same tme, greater number of chldren n the households has caused them to spend more on other meals. Ths s probably due to the fact that school gong chldren spend on food at school canteens and they have greater tendency for eatng out between meals at fast food restaurants. Households make choces more frequently when decdng on the expendtures for dnners and other meals as compared to breakfast and lunch. Increasng the varety of food and mprovng the menu may help to spur the growth of the ndustry. The household members have not much choce but to spend on breakfasts and lunches near to or at ther work places n order to save on travelng tme and cost of eatng at home. As such, steps need to be taken to ensure that these meals are of hgh nutrtonal values, safe to consume and reasonably prced. As for dnners and other meals, ncrease the varety of food and mprovng the menu may help to spur the growth of the ndustry. ACKNOWLEDGEMENTS Ths study was a part of the FAMA-IPTA collaboratve research project on Food Consumpton and Demand Analyses. The researchers are grateful for the fnancal support from FAMA for ths study. We would lke to thank the Department of statstcs for provdng detaled samplng lsts. We wsh to thank and acknowledge the respondent households for ther co-operaton durng the survey. The usual dsclamer apples. REFERENCES Alas, R., Mansor, S.A. and Dayang, M.A.A. (2006). Demand Analyss of FAFH Homes n Malaysa. Staff Paper 6/2006. Faculty of Economcs and Management, Unverst Putra Malaysa, 43300, Unverst Putra Malaysa, Selangor, Malaysa. Becker, G.S. (1965). A Theory of Allocaton of Tme. Economcs Journal, 75, Byrne, Patrk J., Oral Capps, Jr. and Saha, A. (1996). Analyss of away from home expendture patterns for U.S. household, Amercan Journal of Agrcultural Economcs, 78(3), Capps, Jr., O. and Park, L.J. (1997). Demand for prepared meals by U.S. households. Amercan Journal of Agrcultural Economcs, 79, Deaton, A. and Irsh, M. (1984). Statstcal model for zero expendture n household budget shares. Journal of Publc Economcs, 23(1-2), Department of Statstcs Malaysa. (2006). Report on household expendture survey Malaysa 2004/2005. Department of Statstcs Malaysa, Putrajaya. Greene, W.H. (2003). Econometrc Analyss (5 th Edn.). Unted States of Amerca: Prentce Hall, New Jersey. Pertanka J. Soc. Sc. & Hum. Vol. 18 (2)

10 Gaz Nurul Islam et al. Hanna, S. and Carter, L. (1986). Interacton effects of demographc characterstcs on food away from home spendng. In K. P. Schnttgrund (Ed.), Proceedngs of the 32 nd Annual Conference of the Amercan Councl on Consumer Interests (pp ). Hemstra, S. and Km, W.G. (1995). Factors affectng expendtures for food away from home n commercal establshment by type of eatng place and meal occason. Hosptalty Research Journal, 19, Ishda, A., Law, S.H. and Ata, Y. (2003). Changes n food consumpton expendture n Malaysa. Agrbusness, 19(1), Lancaster, K. (1971). Consumer Demand. New York: Columba Unversty Press. Lee, H. and Tan, A. (2007). Examnng Malaysan household expendture patterns on food-awayfrom-home. Asan Journal of Agrculture and Development, 4(1), Lee, J.Y. and Brown, M.C. (1986). Food expendtures at home and away from home n the Unted States A swtchng regresson analyss. Revew of Economcs and Statstcs, 68, Ma, H., Huang, J., Fuller, F. and Rozelle, S. (2006). Gettng rch and eatng out: Consumpton of food away from home n Urban Chna. Canadan Journal of Agrcultural Economcs, 54, McCracken, V.A. and Brandt, J.A. (1987). Household consumpton of food away from home: Total expendture and by type of food faclty. Amercan Journal of Agrcultural Economcs, 69(2), McCracken, Vck, A. and Jon, A. Brandt. (1990). Tme value and ts mpact on household food expendtures away from home. Home Economcs Research Journal, 18, McDonald, J. and Mofftt, R. (1980). The uses of Tobt analyss. Revew of Economcs and Statstcs, 62, Mnstry of Fnance Malaysa. (2004). Economc Report 2004/2005. Kuala Lumpur, Malaysa. Nayga, R. and O. Capps, Jr. (1992). Determnants of food away from home consumpton: Update. Agrbusness, 8, Prochaska, Fred J. and Schrmper, R. A. (1973). Opportunty cost of tme and other socoeconomc effects on away from home food consumpton. Amercan Journal of Agrcultural Economcs, 55(4), Reynolds, A. and Goddard, E. (1993). Food consumpton away from home by type of faclty and type of meal. Workng Paper WP93/10. Guelph: Unversty of Guelph, Department of Agrcultural Economcs and Busness. Res, C. P., Klne, K. and Weaver, S. (1987). Impact of commercal eatng on nutrent adequacy. Journal of Amercan Detetcs Assocaton, 87, Soberon-Ferrer, H. and Dards, R. (1991). Determnants of household expendtures for servces. Journal of Consumer Research, 17(March), Stewart, H., Blsard, N., Bhuyan, S. and Nayga, Jr. R.M. (2004). The Demand for Food Away From Home: Full-Servce or Fast Food? Agrcultural Economc Report Number 829, Economc Research Servce, Unted States Department of Agrculture-Economc Research Servce. Yen, Steven T. (1993). Workng wves and food away from home: The Box-Cox double hurdles. Amercan Journal of Agrcultural Economcs 75(4), Pertanka J. Soc. Sc. & Hum. Vol. 18 (2) 2010