Investigation of factors affecting consumers bread wastage

Transcription

1 Journal of Agrcultural Economcs and Development Vol. 2(6), pp , June 2013 Avalable onlne at ISSN Academe Research Journals Full Length Research Paper Investgaton of factors affectng consumers bread wastage Shahnoush, N. 1, Saghaan, S. 2, Reed, M. 2, Froozzare, A. 1 and Jalerajab, M. 3 1 Department of Agrcultural Economcs, Ferdows Unverst of Mashhad, Iran. 2 Department of Agrcultural Economcs, Unverst of Kentuc, USA. 3 Agrcultural Economcs, Azad Unverst, Tehran, Iran. Accepted 1 Ma, 2013 The am of ths paper s to nvestgate factors affectng bread wastage. We appled the Ordered Logt Model to cross-sectonal data of Iranan households. Households were dvded nto three groups based on ther measures of bread wastage. The results ndcate that the number of vsts to the baer per wee, means of gong to the baer, method of carrng bread, watng tme n the baer, assessment of bread prce and qualt, monthl household ncome, household educaton, household average age, average rate of bread consumpton, preference for consumng fresh bread, method of preservng bread, mother s educaton, mother s job, and number of household members, all nfluence measures of household bread wastage. Ke words: Ordered Logt Model, consumpton, bread wastage, Iran. INTRODUCTION The world experenced a sharp ncrease n food prces durng 2008/2009 and there has been a recent recurrence n 2011/2012 (Fgure 1). The causes of these food prce spes are currentl beng debated among economsts (Heade, 2011; Mason et al., 2011; Pesse and Thrtle, 2009) and t s partcularl mportant to separate short-term factors (such as perodc droughts and one-tme events) from long-term secular factors (such as ncreased ncomes n emergng countres, hgher populaton, and clmate change). Once the professon sorts out the fundamental (and lel long-run) movements of food prces, polc maers can begn to mae better decsons on was to mprove food avalablt and reduce scarct. Most observers beleve that the world s n a new envronment where food prces wll be generall hgher and more volatle so that recent ears wll be more ndcatve of the future. The emphass on food securt n the formulaton of polces b man governments shows that there s serous concern on the future avalablt of grans and other basc foodstuffs to provde for health lvng. Even f these projectons of hgher food prces n the future are not true, t s clear that there are some food polces that are not encouragng effcentl use of the producton that taes place. There s sgnfcant pressure on governments to change polces that encourage neffcent use of food products (ncludng food as a fuel) so that the avalable food supples are consumed n the hghest and best use. Some food subsdes provded b governments, partcularl those that are not targeted to need famles, are under ncreased scrutn because the are ver costl and could promote wastage and overconsumpton. The basc dea behnd subsdes for basc food tems s rather smple - t allows households on ver low ncomes to purchase vtal foods and food ngredents at affordable prces. It s a socal safet net that s relatvel eas to admnster and effectve n eepng the poorest households from severe deprvaton. There are normall Correspondng author. E-mal: mrreed@u.edu. Tel:

2 J. Agrc. Econ. Dev. 247 Fgure 1. FAO Food Prce Index, , Nomnal and Real. Source: FAO (2012). leaages to hgher ncome households that are not economcall stressed so there ma be qualfcaton requrements that target the subsdes to ver poor famles, but ths s not alwas the case. Furthermore, hghl subsdzed products to ver poor famles could be sold to hgher ncome famles to undermne the purposes of the subsdes. These ssues are mportant for determnng food subsd polces, but the are not relevant to the current analss. Ths stud deals specfcall wth factors that affect bread wastage through a surve of Iranan households. Bread subsdes alone have accounted for 60% of Iranan food subsdes recentl (totalng about 8 trllon rals or $800 mllon) and the have a ver large mpact on the retal prce of bread (through flour prce subsdes to baers) (Bread subsdes were elmnated n 2010, wth protests b consumers. There was substantal poltcal pressure on the Iranan government to rensttute bread subsdes to mprove food securt. Currentl there are cash subsdes pad to low ncome households). The prce of bread s so low that there s an entre salvage ndustr that collects stale bread from households for use n anmal feed and other purposes. Yet bread also accounts for 47% of the calores consumed b Iranans (Mnstr of Agrculture). The objectve of ths paper s to nvestgate the effects of household soco-economc factors and demographcs, bread purchasng patterns, and food culture on bread wastage. Food wastage can tae place n man was (ncludng the use of food n relatvel low-valued forms such as anmal feed or even fuel), but for the purposes of ths analss t s the dscardng of food so that t s not consumed b humans. Hghl subsdzed food prces wll be a contrbutng factor because subsdes change the prce of food relatve to other economc factors (such as wages and preparaton costs) and dstort behavor toward treatng food as beng less valuable than other consumer products. In some countres, the prce of subsstence goods (due to government subsdes) s so low that consumers do not care whether some of the products spol and are dscarded before the are consumed. The major cost nvolved n obtanng the product s the trp to the store (and an queung) and that can be qute low f a chld does the shoppng. Ths paper loos at the degree that subsdzed food products, partcularl subsdzed bread prces n Iran, encourage all households to dscard food and allow ts wastage. The specfc objectves are to provde evdence on bread wastage n Iran and nvestgate the effects of factors that ncrease (or reduce) bread wastage b Iranan households. The analss also suggests was that the government can mprove bread utlzaton whle stll mantanng low bread prces for the poor. Wheat, as the man food for half of the world s populaton, s vtall mportant as a food source. Ths crop s cultvated for bread, anmal feed, and ndustral consumpton throughout the world. For Iran, wheat s consdered the most mportant food crop and nutrent; t accounts for more cultvated area and consumpton than an other food product. The Research Plannng Insttute estmates that Iran needs almost 12 mllon tons of wheat per ear to fulfll normal consumer needs (Mnstr of Agrculture). Iran has been able to produce a large amount of wheat b ncreasng ts elds, but man experts beleve that wheat consumpton wll outpace producton n the long run. Thus, t s lel that Iran wll need to mport wheat n the future and use precous hard

3 Shahnoush et al. 248 currenc n order to pa for these mports. Ths puts ncreasng pressure to nvestgate was to use wheat more effcentl and to reduce wastage, especall bread waste. Thus, ths stud nvestgates factors that affect bread wastage n the consumpton process. BACKGROUND Bread wastage n Iran has been extensvel nvestgated n the lterature; however, most studes concentrate on the techncal features of bread producton, such as dough preparaton, gluten percentage, coong tmes, oven effcences, etc. Descrptve research methods are usuall appled to nvestgate soco-economc factors affectng bread wastage (Abed-Shapoorabad, 2005; Azz, 2004; Iran, 2004; Iran and Yazd-Samad, 2005; Zare and Shearforush, 2004; Azadbaht et al., 2007). The two studes most related to ths research are Aut et al. (2003) and Abed-Shapoorabad (1996). Aut et al. (2003) studed the effects of major socoeconomc characterstcs of households on bread consumpton and bread wastage n urban areas of Adana provnce, Ture. The found that 4.45 loaves of bread were consumed per household per da, whle 4.65 loaves were purchased per household per da. As household ncome ncreases, the amount of bread consumpton decreases, as s expected, ndcatng that bread s an nferor good. The found that the average rate of dal bread wastage of the households analzed was 9.63%. Bread wastage was hghest n the hghest ncome group and smallest n the lowest ncome group. Abed-Shapoorabad (1996) studed factors affectng the amount of bread waste usng a multvarate lnear model for one provnce n Iran. He found that the number of people purchasng bread, number of tmes of gong to the baer, tme spent bung bread, and educaton level of household head were the most mportant factors affectng bread wastage. These are some of the varables that are used n the present stud. METHODOLOGY Bread wastage s not readl observable and there s no credble government data; thus a household surve s the onl reasonable approach to address the stud objectves. The lterature on bread wastage n Iran was revewed (there s an extensve lterature because of the long-runnng bread subsdes) to determne varables that should be ncluded n the surve. A team of scentsts (ncludng economsts, other socal scentsts, and food scentsts), government offcals, and baers also met to construct the surve and mae sure that all the relevant factors nfluencng bread wastage were ncluded n the surve. Households and baers n Mashhad, Iran were also ntervewed to mae sure that the surve was comprehensve and useful for the analss. The populaton varance assocated wth bread wastage was unnown so the necessar sample sze could not be determned a pror. The surve was plot tested wth 160 households and the results from that plot were used to estmate the populaton varance. Ths varance estmate was nserted nto the formula for optmal sample sze (1,409 n ths case) n order to determne the number of households that were surveed (Gujarat, 1987). The sample was drawn from a smple random samplng of 1409 Iranan households n 2008 from all of the major ctes n Iran. Students were sent throughout the countr (after the were traned regardng the surve and ntervew technques) to conduct the ntervews n person. The surve was constructed wth ntervals for bread wastage because household members have a dffcult tme n provdng an exact number for wastage, but are comfortable n assgnng the wastage wthn an nterval. The surve found that 65% of Iranan households wasted more than 200 g of bread per capta per wee; 13% of households had per capta bread wastage of more than 1.83 g (a ver large number). Nearl one-half (48%) of the Iranan respondents felt that ther bread wastage was n the lower two categores, though. The now about the maret for stale bread, though, wth 62% sang that the sell ther wasted bread. Iranans love fresh bread and 44% of the respondents sad ther bread had to be ver fresh (44% sad the dd not care much). Almost 75% of bread buers waled to the baer and 68% sad the baer was close (on average wthn sx mnutes for them). Those who sad the baer was close vsted the baer more than fve tmes per wee. The rest of the respondents sad the baer was dstant and t too them an average of seventeen mnutes to get to the baer. The stll vsted the baer an average of four tmes per wee. Most respondents (74%) beleved that the qualt of bread s n the lowest categor. A lttle more than one-half of the respondents (53%) beleved that the poor bread qualt s the reason for bread wastage and 31% sad the onl wa to reduce bread wastage s to ncrease bread qualt. There seems to be a msmatch between the bread that Iranans want (the tradtonal thn bread) and the thc bread (that s soft n the mddle) that s suppled b most baeres. Sxt percent of households purchase the thc bread, whle 45% of households report the cannot get the tpe of bread the desre. Less than 10% of the respondents thought that the prce of bread was too low. The nvestgaton of the factors that have mpact on household bread wastage s a classc case of an Ordered Logt Model. The Ordered Logt Model assumes that there s a contnuous process explanng an unnown varable to ndependent varables x b some functon. In the logt model, addtve error terms are used, so that the underlng process s gven b:

4 J. Agrc. Econ. Dev. 249 x where, (1) s a contnuous varable of per capta bread waste, the βs are parameters to be estmated, s vector of ndependent varables ndcatng characterstcs of household, and s the dsturbance term, assumed to be ndependent across observatons; can tae an value and the subscrpt refers to the household number. Remember that s not observable so standard regresson technques are not sutable for estmaton. If the observed varable s gven n ordered classes (1, 2,..., J), the relaton between and ntervals nformaton... J f f f... f of s that adjacent correspond wth qualtatve. Ths relaton s gven b: 1 2 J 1...,,, 2 3 1, 1,..., n, 1,..., n, 1,..., n,... 1,..., n Parameters and the thresholds,..., ) are (2) ( 1 J 1 smultaneousl estmated usng the maxmum lelhood method, whch maxmzes the probablt of correct classfcatons. We calculate the probablt (Pr) that J b: Pr( J) Pr( J 1 F( x ) Pr( J 1 ) n1 x ) To meet the requrements of a probablt model (monotonc-ncreasng CDF wth results lng between 0 and 1), the dsturbances are assumed to possess a logstc dstrbuton, leadng to a cumulatve logstc transformaton functon L. Ths functon maps the admssble area of, that s, (, ) to [0,1], wth a frst dervatve that s alwas postve (Sonneveld and Dent, 2007). Ths model s defned as the followng: j ( x) log j j ( x ) Where j x x... x, j 1,..., J ; 1,..., n j s cumulatve probablt, that s, ( x ) x P j x j. Estmaton of parameters n ths model should meet the regresson hpothess. Interpretaton of the coeffcent n the Ordered Logt Model s not drect; hence, n order to nterpret them, the margnal effects were used. Margnal effect of one unt varaton n x on probablt of group j s: P j x x x x j x x j 1 x Where j ( x ) j ( x ), x 0 and j. It s necessar to menton that margnal effect for a bnar varable s calculated ndrectl and t s equal to the dfference between the probabltes of two possble states. RESULTS As mentoned earler, ths stud apples the Ordered Logt Model to determne effectve factors on the level of household bread wastage. Table 1 shows the varables ncluded, ther defnton and ther expected nfluence on bread wastage. The SPSS 11.5 software pacage was used to estmate the Ordered Logt Model. Prelmnar data analss ndcated that average per capta bread waste of Iranan households vared wdel among household groups. The Tue HSD test was used to dvde households nto three groups based on ther per capta bread waste measures. The groups are: low and mddle (Group 1), hgh (Group 2), and ver hgh (Group 3), and households are dvded nto these three categores. The results of applng the Ordered Logt Model to nvestgate the effectve factors on a household n each of these three groups are presented n Table 2. The Wald statstc for each coeffcent s used to judge sgnfcance for coeffcents wth an Ordered Logt Model. Based on 2 R pseudo statstcs, t can be seen that the estmated Ordered Logt Model s a sutable regresson and that the ndependent varables of the model explan the varaton n bread wastage from dfferent household groups. Moreover, the model predcts the correct categor for bread wastage 71% of the tme, ndcatng that t has good forecastng ablt. In addton, the results of the parallel regresson shown n Table 3 specf that the assumpton of proportonal odds among the three groups s vald. The Pearson and Devance goodness of ft measures show no ll-fttng tendences n the model (Table 4). Hence, based on all of these crtera, ths model s approprate and trustworth. It s clear from Table 2 that the means of gong to the baer, the method of carrng bread, the number of j j1 x

5 Shahnoush et al. 250 Table 1. Defnton of appled varables n the Ordered Logt Model. Varable Defnton Expected sgn Bread purchaser Cooperaton=1; ndvdual=0 - Dstance to the baer Far and average=1; near=0 + Means of gong to the baer Walng=1; a vehcle=0 + Method of carrng bread On hand=1; wth a baset, cloth=0 + Cool bread after purchasng? Yes=1; No=0 + Number of household members (contnuous) -/+ Number of trps to the baer per wee (contnuous) -/+ Frequenc of eatng bread per wee (contnuous) - Preference for fresh bread Low and mddle =1; hgh =0 - Purchase more than dal need? Yes=1; No=0 + Method of preservng bread In refrgerator and freezer=1; other=0 - Watng tme at baer (contnuous) mnute + Assessment of bread qualt Good=1; mddle and wea=0 - Assessment of bread prce Low=1; mddle and hgh=0 + Assessment of bread s salvage value Mddle and hgh=1; low=0 + Monthl household ncome (contnuous) 100,000 Ral + Educaton level for household head Illterate=0; elementar=1; gudance=2; dploma=3; - Mother s educaton upper dploma=4; lcense=5; master=6; PhD=7 - Mother s job Houseeeper=1; emploer=0 - Average household age (contnuous) ear -/+ Average rate of food consumpton (contnuous) -/+ Average rate of bread consumpton (contnuous) -/+ Household bread waste Mddle-low=1; hgh=2; ver hgh=3 Table 2. Results of the Ordered Logt Model. Varable Coeffcent Standard error Wald Bread purchaser Dstance to the baer Means of gong to the baer Method of carrng bread Cool bread after purchasng Number of household members Number of trps to the baer per wee Frequenc of eatng bread per wee Preference for fresh bread Purchase more than dal need? Method of preservng bread Watng tme at baer Assessment of bread qualt Assessment of bread prce Assessment of bread s salvage value Per month ncome of household Educaton level for household head Mother s educaton Mothers job Average household age Average rate of food consumpton

6 J. Agrc. Econ. Dev. 251 Table 2 Contd. Average rate of bread consumpton Percentage of rght predcton 70.25% Pseudo R-Square (Alternatves of Pseudo R-Square recommended b Ben Ava and Lerman (1985) and Pa and Saleh (2008)) Cox and Snell 0.16 Nagelere 0.19 McFadden 0.11 Dependent varable: Ordered varable household bread waste groups. Sgnfcantl dfferent from zero at the 1% level; Sgnfcantl dfferent from zero at the 5% level; Sgnfcantl dfferent from zero at the 10% level; t-rato s greater than 1.0. Table 3. Results of parallel regresson test. Model -2 Log Lelhood Ch-Square Level of sgnfcance Current model General model Table 4. Indces of goodness of ft. Crteron Ch-Square Level of sgnfcance Pearson Devance vsts to the baer per wee, the watng tme n the baer, the assessment of bread prce, monthl household ncome, educaton level for the head of household, average age of household, and the average rate of bread consumpton postvel affect household bread waste. In other words, ncreasng these ndependent varables ncrease the probablt that a household wll be n the hgher bread waste groups. Furthermore, preference for consumng fresh bread, method of preservng bread, assessment of bread qualt, the mother s educaton level, the number of household members, the mother s job and the average rate of food consumpton negatvel affect household bread waste. In other words, ncreasng these ndependent varables ncreases the probablt of embeddng households n the lower bread waste group. As stated earler, nterpretaton of coeffcents for the Ordered Logt Model s not possble. Hence, t s necessar to estmate margnal effects, whch are reported n Table 5 for ths model. The margnal effects of the Ordered Logt Model show that walng to the baeres - nstead of usng a vehcle - carrng bread b hand and wthout a cover, ncreasng the number of trps to the baer per wee, ncreasng watng tme at the baer, hgh preference for consumng fresh bread, low assessment of bread prce, ncreasng monthl household ncome, ncreasng the educaton level of the household head, ncreasng the average age of the household, and ncreasng the rate of bread consumpton, all decrease the probablt of ncludng a household n the low-mddle waste bread group (or ncrease the probablt of placng a household n the hgh and ver hgh waste groups). Based on the results of ths stud, cooperaton n bread purchasng, preservng bread n a refrgerator, good assessment of consumpton bread qualt, ncreasng the mother s educaton, ncreasng the number of household members, havng the mother as houseeeper, and ncreasng the average rate of food consumpton all ncrease the probablt of placng a household n the lowmddle waste bread group. On the other hand, these varables decrease the probablt of ncludng a household n the hgh and ver hgh waste groups. When there s onl one purchaser of bread, the household reduces the number of trps to the baer and purchases more bread per trp. Ths decreases the probablt of ncludng the household n the low-mddle bread waste group. Walng to the baer ncreases the probablt of ncludng the household n the hgh and ver

7 Shahnoush et al. 252 Table 5. Margnal effects of the three groups of households n terms of ther bread waste. Varable Margnal effect of Group 1 Margnal effect of Group 2 Margnal effect of Group 3 Bread purchaser Dstance to baer Means of gong to the baer Cool bread after purchasng? Method of carrng bread Number of trps to the baer per wee Preference for fresh bread Method of preservng bread Watng tme n the baer Assessment of bread qualt Assessment of bread prce Assessment of bread s salvage value Monthl household ncome Educaton level for household head Mother s educaton Mothers job Number of household members Average household age Average rate of food consumpton Average rate of bread consumpton hgh bread waste groups because people do not le walng, so the tr to bu bread from the nearest baer wthout pang attenton to the taste and qualt of bread. If the bread s not good, t s more lel to be dscarded. Carrng bread n a baset, cloth, or other contaner ncreases the probablt of ncludng a household n the low-mddle bread waste group because these contaners mantan the qualt of bread. Increasng the number of trps to the baer each wee ncreases the probablt of a household that s n the hgh and ver hgh bread waste groups because Iranans usuall bu more bread than ther dal need. The prefer to eat fresh bread, so f the go to the baer often, the are more lel to dscard older bread, brngng about more waste. Preservng bread n the refrgerator ncreases the probablt of ncludng a household n the mddle-low bread waste group and decreases the probablt of ncludng the household n the hgh and ver hgh bread waste groups because refrgeraton decreases bactera actvt and preserves bread qualt. Increased watng tme at the baer encourages households to bu more bread than ther dal need, resultng n an ncreased probablt that the household s n the hgh and ver hgh bread waste groups. If the household vews the prce of bread as low, due to the subsd or other reasons, there s an ncreased probablt that the household wll be n the hgh or ver hgh bread wastage groups. A low salvage value for bread reduces the opportunt cost of waste and therefore, ncreases the probablt of ncludng a household n the hgh and ver hgh bread wastage groups. As llustrated n Table 5, ncreasng the per capta ncome of a household reduces the real value of bread, so the household does not pa suffcent attenton to use bread effcentl; ths leads to a decreasng probablt of ncludng the household n the mddle-low bread wastage group. Hgher educaton for the household head brngs about more dscrmnatng atttudes to bread qualt and ts nutrent values. Hgher educaton also ncreases per capta household ncome. Consequentl, these factors ncrease the probablt of ncludng the household n the hgh and ver hgh bread waste groups and decrease the probablt of ncludng t n the mddle-low bread waste group. Due to the manageral role of the mother, ths stuaton s reverse n the case of the mother s educaton level; a more hghl educated mother reduces bread wastage. However, mothers emploed outsde the home do not have as much tme to manage and admnster ther home envronment, whch ncreases the probablt of ncludng a household n the hgh and ver hgh bread waste groups. Increasng the number of household members ncreases the probablt of ncludng t n the mddle-low bread wastage group. Ths ma be because larger households are more tradtonal and the place a hgher

8 J. Agrc. Econ. Dev. 253 ntrnsc value on bread. The also have more chldren and the mght be able to fnd more was to mae bread consumpton appealng to ds. Fnall, the are more lel to have a lower per capta ncome, mang them more cognzant of bread wastage. Increasng the average household age decreases the probablt of ncludng a household n the mddle-low bread wastage group, and ncreases the probablt of ncludng a household n the hgh and ver hgh bread wastage groups. Households wth older adults have man problems wth bung bread so the mght purchase more bread durng each vst to the baer, but the also have more problems n eatng stale bread. Lastl, ncreasng the average rate of food consumpton ncreases the probablt of ncludng a household n the mddle-low bread wastage group on the grounds that these tpe of households tpcall have a lower level of per capta ncome and spend more of ther ncome on food. Increasng the average rate of bread consumpton decreases the probablt of ncludng a household n the mddle-low bread wastage group and ncreases the probablt of ncludng them n the hgh and ver hgh bread wastage groups snce the consume more bread and produce more bread waste. Conclusons Ths stud fnds that there s sgnfcant bread wastage b Iranan households, whch s not surprsng gven the hghl subsdzed prce for bread. Such a low bread prce encourages consumers to waste t and also dscourages baers from provdng hgh qualt bread. Ths encourages hgh levels of bread wastage. The Iranan government ma fnd t dffcult to ncrease bread prces much because of the need to mprove food securt for the poor. Bread wastage s not a problem wth low ncome consumers; t s much hgher for households wth more educaton and hgher ncomes. Hgher ncome, more educated households realze that bread qualt (and nutrtve value) can deterorate qucl over tme so the bu bread more often and dscard bread more readl. The readl provde ther stale bread to recclers to tae t awa. There s a role for ncreased educaton of bread producers and consumers on practces that wll reduce bread wastage. Baers can be better educated so the handle bread and flour n was that mprove qualt. The also need to pa better attenton to the tpe of bread desred b households (the thn tradtonal bread) and mprove ther bang slls for the thcer bread. The low prce of bread provdes lttle ncentve for the baer to worr about these ssues, though. Consumers need to understand that the thc bread provded b baers s more nutrtous than the tradtonal bread and t needs to be preserved to mantan qualt. Placng the thc bread n the refrgerator wll preserve ts qualt for a much longer tme. Coolng the bread properl and placng t n a wrapper (rather than smpl carrng t b hand) wll also ncrease the bread s shelf lfe. Better preservaton methods wll reduce the number of trps to the baer too. The mother s admnstratve role s e to educatng households on mproved bread utlzaton. Usng popular meda to educate mothers on bread nutrton, handlng, and preservaton (ncludng refrgeraton and postve characterstcs of thcer bread) wll decrease per capta bread wastage and enhance the household s det. Ths wll be a challenge because cultural tradtons and household norms are frml establshed. The tradtonal thn bread s consumed (though t has lttle nutrtonal value) and t s not preserved because t s qute cheap and s easl dscarded. Reducng bread wastage s an uphll battle wth current bread subsdes, but t s a battle that should be fought b the authortes. REFERENCES Abed-Shapoorabad A (2005). Surveng Bread Waste n Shahrord. Second Conference of Agrcultural Products Waste. Tehran. Aut G, Hlal I, Tufan B, Sertac O (2003). Bread Consumpton and Waste of Households n Urban Area of Adana Provnce. Electronc Journal of Polsh Agrcultural Unverstes, Food Sc. Technol., 6(2). Azadbaht N, Nasr Z, Rashd R, Khosravnejad S, Sahpoosh S (2007). Investgaton of Reasons and Measures of Bread Waste n Lorestan. Thrd Conference of Agrcultural Products Conference. Tehran. Azz M (2004). Surveng Methods of Reducng Bread Waste and Increasng ts Qualt. Frst Conference of Preventve Methods of Wastng Natonal Resources. Tehran. Ben-Ava, M, Lerman S (1985). Dscrete Choce Analss. MIT Press, Cambrdge, MA. Food and Agrculture Organzaton database ( tuaton/wfs-home/foodprcesndex/en/) Gujarat D (1987). Basc Econometrcs. McGraw-Hll, New Yor, N.Y. Heade D (2011). Rethnng the Global Food Crss: The Role of Trade Shocs. Food Pol., 36: Iran P (2004). Surveng Reasons of Dfferent Bread and Flour Waste. Frst Conference of Preventve Methods of Wastng Natonal Resources. Tehran. Iran P, Yazd-Samad B (2005). Changng Producton and Consumpton Models of New Breads n Order to Prevent Waste. Second Conference of Preventve Methods of Wastng Natonal Resources. Tehran. Mason N, Jane T, Chapoto A, Donovan C (2011). Puttng the 2007/2008 Global Food Crss n Longer- Term Perspectve: Trends n Staple Food Affordablt n Urban Zamba and Kena. Food Pol., 36: Mnstr of Agrculture, Agrcultural Statstcs. Iran.

9 Shahnoush et al. 254 Varous Issues. Pa C, Saleh W (2008). Modelng Motorcclst Injur Severt b Varous Crash Tpes at T-Junctons n the UK. Safet Sc., 46: Pesse J, Thrtle C (2009). Three Bubbles and a Panc: An Explanator Revew of Recent Food Commodt Prce Events. Food Pol., 34: Sonneveld B, Dent D (2007). How Good s GLASOD? J. Envron. Manage., 20: Zare M, Shearforush S (2004). Optmal Usage of Bread Waste and Publc Health. Frst Conference of Preventve Methods of Wastng Natonal Resources. Tehran.