Report No. 39738-YE Yemen Towards Qat Demand Reduction June 2007 Country Department III Sustainable Development Department Middle East and North Africa Region Document of the World Bank Abstract................................................................................................................................. Table of Contents Acknowledgement................................................................................................................. i ExecutiveSummary............................................................................................................. ..i I Introduction.................................................................................................................. 11 1 I1 MainFindingsofthe Qat Consumer Survey .. ............................................................... A. Qat Consumptionis Widespread.............................................................................. 5 B. Reasonsfor ChewingandAbstaining....................................................................... 5 9 C Qat Use Adversely Affects Healthand Family Life . ................................................ D. QatUse DrainsProductivityandFuelsPoverty..................................................... 11 13 I11 DiscouragingQat Consumption ................................................................................ . E. ConsumersNeedGovernment Help To Kickthe Habit .......................................... 17 20 A. ImplementingPastPolicy Decisions...................................................................... B. Directionsfor FuturePolicy Dialogue.................................................................... 20 21 References .......................................................................................................................... 28 Annex 1:The QatConsumer Survey Sample........................................................................................ Annexes: 29 Annex 2: The RelationshipBetweenAge andQat ChewingFrequency for MenandWomen Annex 3:HouseholdExpenditureon Qat.............................................................................................. ............31 32 Boxes Box 1: Analysing Trends inQat Expenditure .................................................................... Box 2: PrinciplesofDrugDemandReduction................................................................ . 4 22 Box 3: OECDTobaccoConsumption:FromGlamourto Social Stigma ................................... 23 Table1: Sample PopulationChewingqat by Gender andGovernorate.................................................. Tables 5 Table2: Age ofFirstQat Use by Region ............................................................................................... Table3: Qat ChewingLocation.............................................................................................................. 8 9 10 Table5: ReasonsWhy ConsumersUse Qat......................................................................................... Table4: ReasonsWhy Consumers Start to Chew qat.......................................................................... 10 Table6: ReasonsWhy ConsumersAbstainfrom Qat.......................................................................... Table7: Consumer PerceptionsRegardingtheHealthEffectsof qat .................................................. 11 12 Table 9: DurationofDailyQatChewingbyRegion............................................................................ Table 8: Ways Consumers TreatAdverse HealthEffects ofQat ......................................................... 13 Table 10:Effect ofQat on ImmediateWork Performance.................................................................... 14 15 Table 12: PopulationWho Cannot AffordQat andMeet Their BasicNeeds ....................................... Table 11:Effect of QatonDay-AfterWork Performance.................................................................... 15 16 Table 13: What would respondentspurchase ifthey hadmorefunds?................................................. 17 Table 14: ConsumerPerceptionsofQatChewing................................................................................ Table 15: Shouldthe GovernmentLimit QatChewing?....................................................................... 17 18 Table 16: Respondent's Suggestionsfor PossibleAlternatives to Qat.................................................. 19 Figures Figure 1: ChewingFrequencyof YemeniMales .................................................................... 6 Figure2: ChewingFrequency of Yemeni Females ................................................................ 6 Figure 3: RegionalDifferences inMen's Chewing ................................................................. 7 Figure4: RegionalDifferences inMen's Chewing................................................................. 8 Figure 5: Amount of Time Spent ChewingQat PerDay ....................................................... 14 Abstract This report, based on a household survey conducted in 2006, discusses options for discouragingqat consumptionin Yemen. It draws on a survey-the first representative data collection exercise aimed specifically at assessing the qat consumptionphenomena-which confirms that the use of this drug is widespread. Qat is consumed by men, women and children; its use is extremelytime consuming; it drains the family budget; has adverse health effects; negatively affects work performance and thus contributes to poverty. Weaning consumers from the qat habit will be difficult, because its production accounts for some 6 percent of GDP and 14 percent of total employment. Qat consumptionrequires around 10 percent of the household budget of all income groups, which comes at the expense of basic food, education and health. To reduce qat consumption, this note recommends a set of economic and non-economic policy measures. These include: increasing the tax burden; buildingpublic awareness; incorporatingtraining on the hazards of qat inthe school system; enforcingpublic policies aimed at discouragingqat consumption(e.g., extension of working hours); closingknowledge gaps anddevelopingviablecrop diversificationprograms. Acknowledgement This paper was preparedby Peer Gatter (consultant), Mustapha Rouis (task-team leader) and Steven Tabor (consultant). Peer Gatter provided a first draft based on a field survey conducted for the Bank under his supervision with assistance from Khaled al-Hariri and Abdul Ghaleel al-Hemiary (consultants). Naji Abu-Hatim provided guidance on the survey design and feedback on its findings. The authors would like to acknowledge the comments and suggestions received from Inger Andersen, Meskerem Brhane, Yasser El-Gammal, Daniela Gressani, Vijay Jagannathan, Michal Rutkowski, Trayambkeshwar P. N. Sinha, Thirumalai G. Srinivasan., Gaiv Tata (all Bank staff), Christopher Ward (consultant and peer reviewer), and Dr.YahyaAl-Mutawakel(Government ofYemen) Vice President Daniela Gressani Country Director EmmanuelMbi Sector Director Inger Andersen Task Manager MustaphaRouis i Yemen Towards Qat Demand Reduction ExecutiveSummary i.Qat(Cathaedulis)playsamajoreconomicroleintheYemenieconomy. Itaccountsfor around 6 percent of GDP, 10 percent of consumption, one-third of agricultural GDP, and provides employmentfor one inevery seven working Yemeni. As the predominant cash crop, the income it generates plays avital role inruraleconomies. But it also depletes scarce water resources and has crowdedout production of essential food crops and agriculture exports. ii.Untilthe 1960s,qatchewingwasanoccasionalpastime,mainlyfortherich. Now, itis chewed several days a week by a large share of Yemen's population. Widespread qat consumption has grave consequences: its use is linked to widespread child malnutrition and household food insecurity since spending on it pre-empts expenditures on basic foods and essential medicines. The adverse health effects of qat are many and include high blood- pressure, under-weight children (when pregnant women chew qat), cancer (from consuming pesticide residues), and dental diseases. Consumers spend, on average, nearly 10 percent of their income on it, and the physical act of using the drug requires several hours ina day. The culture of spending extended afternoonhours chewing qat is inimicalto the development of a productive work force, with as much as one-quarter of usable working hours allocated to qat chewing. iii.Discouragingqatuseisacomplextask.Inrecentyears,theGovernmenthastakenbold steps to initiate a debate on qat (notably the 2002 national conference) and to encourage cultivation of substitute crops. Despite these measures, qat production and consumption continues to rise at a blistering pace. A better understanding of the extent and nature of qat consumption is needed if the Government is to formulate effective policies aimed at encouraging consumers to reduce qat use. What is known about qat consumption is often drawn from small samples and anecdotal observation. To help fill this knowledge gap, a sample survey of qat consumptionhabits was conducted by the World Bank inApril-May of 2006. The survey covered 4,027 Yemenis above the age of 12 in seven of Yemen's 21 Governorates, including the Sana'a Capital Secretariat, Sana'a Governorate, al-Hodayda, Ta'izz, Marib, Aden andHadhramawt. iv. The survey results show that qat is a nationwide habit. Some 72 percent of Yemeni males reported that they chew qat, compared to 33 percent of females. Most Yemeni's are habitual users-more than halfofthose who chew do so each day of the week. Moreover, qat use incidence is highest in those parts of the country where it is cultivated or where the consumptionhabit has a long history. v. Yemenis start to consume qat during their teenage years, with first use occurring on average between 16 and 24 years of age. Men tend to use qat until they reach old-age, at which point consumptiontapers off (as other illnesses set in). By contrast, women tend not to use qat before they are married, and the older they get, the more likely they are to use it. Qat i s predominatelychewed inthe home, and some 75 percent of qat users declare that they chew during their leisure hours. Menare more likely to relax and chew (82 percent), while some 44 percent of the women combinework and chewing. ii vi. Most consumers start their qat habit because o f peer pressure: Some 66 percent o fthe men and 72 percent o f the women claim that the reason they started to chew qat i s because "all my friends andfamily chew". Those who abstain do so because they are aware o f the adverse health effects and the highcost o f usingqat. vii. Users do know that qat is harmfulto their health. Some 80 percent o f all male users and 70 percent o f all female users report health disorders that are perceived to be related to qat consumption. In addition, a quarter o f all users suffer from chronic sleeplessness that i s related to qat use. Practically all o f the qat users self-treat these ailments with a range o f traditional cures and modern medicines. viii. Qat use hampers productivity and directly contributesto poverty. The enormous amount o f time spent chewing qat is time that, in other countries, is typically spent in productive activities. Some 36 percent o f the users spend 2-4 hours per day; 35 percent spend 4-6 hours a day, and an astonishing 22 percent spend more than 6 hours a day chewing qat. Most users believe that qat enhances their immediate work performance, but about a third reports that work performance i s impaired the day after chewing qat. The overwhelming majority o f qat users are aware that its use i s costly, and that it has a detrimental effect on their ability to afford basic foods, medicines and other essentials. Some 94 percent o f non- chewers and 77 percent o f chewers report that qat has a negative effect on the family budget. Tragically, just under a fifth o f all qat users are forced into debt to finance their chewing habit. ix. Most users believe that qat is bad for them. More than 70 percent o f the respondents describe qat chewing as a "bad habit", that is also bad for the economy and bad for the nation's image. Users want to "kick the habit" but they cannot. Either because o f social pressures, or because o f the psychological dependency resulting from prolonged use, users do not feel that they can stop usingqat on their own. Some 53 percent o f all male and 61 percent o f all female respondents declare that Government intervention is necessary to address the qat problem. x. What can the Government do? The Government can begin to take steps to curb the widespread demand for qat. National leaders need to be engaged in campaigns to reduce qat use, and they can start to show their support by vigorously enforcing the policy measures adopted since 1999 to curb the qat habit. In the future, the Government can help discourage qat use by making it a far more expensive habit. If cigarettes are a fair comparator, then consumption can be taxed at a substantially higher rate, and by shifting from ad valorum to specific taxes and by using tax stamps, those taxes could be far more effectively enforced. Over time, the Government can help to convince consumers that qat use i s a bad habit and should be stopped. Building on the findings from this survey, a well-targeted public awareness campaign could be launched, to let consumers know more about the adverse effects o f qat use, including the fact that most users know it i s a bad habit; the vast bulk suffer adverse health effects; it drains user's budgets and puts them into debt; and that users would like Government help to kick the habit. Information campaigns should initially be targeted to young people, both inside the schools and out, since preventing the qat habit from starting could have a major effect on national demand in a decade or less. Working together with leading NGOs, a combination o f a well-targeted public awareness campaign, peer-to-peer education, and positive examples set by leaders and cultural figures could be used to convince consumers, and especially the youth, that qat use is harmful to themselves and society. ... 111 xi. Government support is also needed to promote effective substitutes for qat consumption. For youth, there i s a pressing need for greater access to sports, youth centers, and cultural activities. For the adult population, the main challenge i s to boost employment demand. Policies aimed at making the economy more competitive and conducive to private initiative will increase the opportunity cost of the tremendous amount of time now devoted to chewing qat. xii. Research on the causes and consequencesof qat consumption, and the behavioural pressures associated with qat uptake is limited in number and scope. While demand- control interventions are being implemented, parallel research activities into these issues may help to refine the targeting of interventions, such as those designed to improve health information for the poor, for maximum effect. More research is also needed to identify an appropriate package of supply-sidemeasures that could be used to mitigate the likely long- term effects of qat cultivation on employment and rural incomes. A combination o f legal reforms, crop diversification measures, improvements in water resource management, and social safety net measureswill be requiredto facilitate the shift in resources that now go into producing qat into other forms o f productive economic activity. iv Yemen Towards Qat DemandReduction I. Introduction 1. Qat is a leafy recreational drug that has been widely grown and used in Yemen for centuries. It i s a mild stimulant whose effects are closely related to those of amphetamines. While users can become psychologically dependant on it, it is not considered an addictive drug. In recent years, rapidly-growing production and use have led to an intensified debate over qat and its effects on Yemeni society. 2. Qat and Society. Until the 1 9 6 0 ~qat chewing was an occasional pastime for the ~ elite. Now, it i s chewed several days a week by a large share of Yemen's population. The big change seems to have been in the 1970s and 1980s due to a rapid increase in incomes, the increased profitability of production for farmers; the introduction of tubewells; the rapid improvements in the rural and main road networks, and the increased leisure-time of those who returnedfrom working overseas. 3. Qat chewing is a deeply engrained custom. It has replaced almost all other forms of relaxation and socialising. Its place in society extends far beyond that of recreational drug use-the role of the qat party in Yemeni life has been variously described as: an emblem of social interaction; a group communion; a manifestation of commitment and conformity; a display of reciprocity; an opportunity for conspicuous consumption; and a venue for social competition (Ward 2000). 4. Qat and the Economy. Qat plays a major role in the Yemeni economy. Qat's contribution to GDP of around 6 percent (2005), including both its direct and indirecteffects, is equivalent to two-thirds of the contribution the oil makes to the economy'. One in every seven working Yemeni are employed in producing and distributing qat, making it the second largest source of employment in the country, exceeding even employment in the public sector2. The distribution network for Qat is undoubtedlythe most advancedinthe nation. 5. Yemen's population of 21 million is predominantly rural (73 per cent) and qat is the single most important cash crop. As a crop, qat is muchappreciatedfor its high profitability, its hardiness and drought resistance, and the few husbandry problems associated with it. Farmers also like the fact that qat can be brought to harvest during most months of the year, that it can be harvestedin small or large quantities accordingto the farmer's need for money, and that it brings cash in on the very day of harvest. Qat is predominately grown by smallholders and the bulk of the revenues generated are retained in countryside. Qat represents only 10 percent of cultivated land3 but generates around a third of agricultural GDP. Nearly one-third of the agricultural labor force is engaged in qat production4. It is exclusively cultivatedinthe highlands and inthe eastern plateau. Recent studies show that qat is 10to 20 times moreprofitable than competing crops (Ward 2000). Ministry ofPlanning and International Cooperation, StatisticalYear Book, 2005. 2004 estimate o f the Ministry of Planning and International Cooperation, Socio Economic Development Plan for PovertyReduction(DPPR). 2005 estimate o f Ministry of Planningand International Cooperation, StatisticalYear Book, 2005 2004 estimate of Ministry of Planning and International Cooperation, Socio-Economic Development Plan for PovertyReduction 6. Qat and PovertyYemen is a large, poor country. In2005, gross national income was estimated at $600 per capita (Atlas methodology) and around 35 percent o f the people were estimated to live in poverty (per 2005 Household Budget Survey estimates). Nearly a fifth o f the children are classified as malnourished and more than 70 percent of the women are illiterate. Qat consumption is one of the key causes of poverty in Yemen. This i s because some 10% o f the household budget is used on it, far exceeding household outlays for basic foods, medicines and other necessities (per 2005 Household Budget Survey, Annex 3). In addition, the time devoted to the daily qat chewing session represents an enormous loss o f potential income, and there are a wide range o f adverse health effects arising from its consumption. 7. While qat cultivation has provided an important source o f rural incomes, it has precluded other and more sustainable forms o f rural economic activity. The area under qat has expanded 13-fold in the last three decades, displacing exportable coffee, fruits and vegetables, sorghum and other foodstuffs. Exports o f other cash crops have fallen while food imports have increased, due to the inroads made by qat inthe rural economy. 8. Qat's role in natural resource use. Qat cultivation has potentially disastrous environmental effects. Being very water intensive5, it reduces not only current production o f alternative crops, but also future production because it is mining non-rechargeable fossil aquifers. Water scarcity in Yemen is serious and getting worse, due to a combination of urbanization and rapid growth inthe numbers o f diesel-powered tubewells used to irrigate qat Per capita availability o f water in Yemen i s at 2 percent of world and 10 percent o f regional averages. Qat growers usually do not own the aquifers they use nor do they pay any fees or charges for such use. They pump water using subsidized electricity and diesel. Already, one- third of all ground water abstraction is utilized for qat cultivation. Groundwater extraction has reached 130 percent o f recharge with agriculture using about 90 percent o f the groundwater and within that; qat i s the leading source o f groundwater depletion. While qat does not require large quantities o f inorganic fertilizer, it does remove considerable amounts o f plant nutrients from the soil, contributing to soil degradation. 9. Social and health effects. Qat use has wide ranging social and health effects. The main psycho-neural effects are said to be wakefulness and suppression o f appetite. Proponents view it as a social convention, promoting harmony and providing a forum for collective decision-making. Many business deals are said to be made duringqat sessions. Opponents say qat is harmful to health, leads to absenteeism and reduced work effort, reinforces gender segregation, keeps men away from their families, and ultimately impoverishes families. The adverse health effects o f qat are many: high blood-pressure, teeth decay, under-weight children (when pregnant women chew qat), constipation, hemorrhoids, hernias, paranoia, and depression. Pesticides used in qat agriculture are a growing cause o f cancer due to consumption o f pesticide residues. Qat use i s also linked to widespread child malnutrition and household food insecurity since spending on it crowds-out household expenditures on basic foods and essential medicines. 10. While the benefits from qat consumption are largely intangible (i.e. fostering social cohesion), the costs certainly are not. On average, consumers allocate nearly a tenth o f their household budget to it, while sample surveys find that poor households allocate as much as 28% o f their income to it (Lenaers and Gatter 2000). The physical act o f using the drug In some areas, mainly inter-mountainous planes, qat cultivation is intensive and heavily irrigated by groundwater. 2 requires several hours ina day. The cultureof spendingextendedafternoonhours chewing qat is inimical to the development of a productive work force, with studies showing that as much as one-quarterof usable working hoursmay be allocatedto qat chewing. 11. Government policy. The Government has adopted a cautious policy stance towards qat. Imports are banned, but production is legal and sales are taxed. A few politicians have spoken out against qat, but there has been no coordinatedanti-qat campaign. Since 1999, the Government has adopted a number of specific policies aimed at controlling, regulating and taxing qat. So far, these policies have failed to deter growth in either production or consumption. But Government policy has succeeded in fostering public debate about qat6, and policy makerscontinueto seek ways to reversethe alarming spread inqat use. 12. Discouraging qat consumption is a critical and complex task. Qat is far too well integratedinto the Yemeni economy and society for significant reduction inits cultivation and use to occur inthe short-term. Drug demandreduction i s best thought of as a long-term task-- -as experiencewith tobacco inthe west illustrates, it cantake decades before large numbers of consumers finally do abandonit. 13. A solid understandingofthe nature of and reasons for qat use is needed ifGovernment i s to take appropriate steps to discourage its use. Unfortunately, rather little is known, outside of anecdotal evidence, about who actually uses qat7. Recently, attempts have been made to analyze national household consumption data to examine patterns in Qat expenditures (see box 1 below). While these findings are instructive, the qat consumption trends identified in the broad national household expenditure surveys appear to suffer from serious under- reporting, making it difficult to reach firm conclusions from surveysthat were not intendedto focus explicitly on qat use. A great deal is known about the production and distribution of qat', and about its health consequences9,but surprisingly, very little is known about the nature and extent of qat chewing; the reasons why consumers do or do not choose to use the drug; who actually uses it; and the influence that qat consumption has within the household, the work place and society at large. 14. A better understanding of qat consumption is needed to inform public policies aimed at reversing growth in demand for this drug. To help close this knowledge gap, a survey of qat consumption patterns, sponsored by World Bank, was undertaken by P. Gatter in mid- See the various public debates on the influence of qat in: Yemen Times: "Will qat be Replaced", 17April 2002; "Poor Yemenis Seek Nirvana with Narcotic Leaf" (March 14, 2003); "qat does not make life a carnival", (30 June 2004); "The Qat Phenomenonand Ideasto Stop its Use" (March 12,2006), "Yemen has great potential so why has it reached this poor state" (12 August 2006); and "Qat Plantations Consume 40% o f Underground y t e r " (20 March 2006). Two Italian doctors studied qat consumption in the 1950s and 1960s in Ta'izz City. Based on 12 years of observations from the Ta'izz hospital, they reportedthat 90 percentof the males over the age of 12 chewedqat, but only 60% were "habitual" chewers. For women, the figures were lower with 59 percent using qat, and only 35 percent consideredto be habitual users. (Mancioli and Parrinello 1967). It would be inappropriateto apply the results oftheir investigationto Yemen as awhole today, given that this study was undertakenin atraditional qat consuming city, the sample was drawn from hospital patients, and its findings are now some forty years out of qate. * The seminalreport on qat production in Yemen is: FA0 (2002): Towards the Formulation of a Comprehensive qat Policy in the Republic of Yemen (Technical Study). Other important studies include those by Amer and Juneid(1996), Milich(1995), Ward et. al. (1998), and Ward and Gatter (2000). The health implications of qat use have been the subject of investigations spanning several decades. Important researchreports on the health implications of qat use include: Alles (1961), Brenneisenet. al. (1990), Halbach (1979), Khalix (1984 and 1987), Kennedy et. al. (1983), Lenaers and Gatter (2000), Lemordant (1983), United Nations (1980), and World Health Organization(1980). 2006. The main findings of that survey are presented in section two of this note. Section three discusses the policy implications of the findings, and sets out a package of policy measures for government to help Yemenis escape the qat habit, and hence to realize their full, drug-free development potential. Box 1: Analyzing Trends in Qat ExpendituresUsingNationalHouseholdExpenditureData Milanovic (2007) has analyzed data from two late-I990s national household expenditures to examine expenditures on Qat in Yemen and Djibouti. The analysis drew on a subset of the questionnaires collected as part o f the 1998 Yemen Household Survey, and for Djibouti, the analysis drew on the 1997 Household Budget Survey. The results confirm that Qat consumption plays an enormous role in the daily life o f households in both countries. Between one-half (in Djibouti) and 70 percent (in Yemen) o f all households report that they have at least one user. The results suggest that qat consumption in Yemen is rather flat across income groups, age and between rural and urban users. In Djibouti, by contrast, the findings suggest that qat consumption increases with income, and appears to substitute for food consumption. Inboth countries, menare found to bemuchmore likely to use qat thanwomen. Caution should be exercised in interpreting the survey results. There appears to be substantial under- reporting o f qat consumption since consumption from home-production is excluded from the survey. In addition, the Yemen survey was launched during a period o f time in which Government was actively discouraging qat consumption. Consequently, respondents may have been reluctant to report on actual qat outlays to government surveyors. Milanovic. Branko. 2007. Oat Exuenditures in Yemen andDjibouti. An Empirical Analvsis. WorldBank. .Mt'tZ/1 Parlet' No. 1425,imred IO.Jcinuar,; 20i17 ccvur'iahkoniitie (It: httu:// mura.ub.uni-muenchen.de/ 1329 ~ 4 11. Main Findingsof the Qat Consumer Survey 15. The qat consumption survey-covering 4,027 Yemenis above the age of 12-was carried out in seven of Yemen's 21 Governorates. These included the Sana'a Capital Secretariat, Sana'a Governorate, al-Hodayda, Ta'izz, Marib, Aden and Hadhramawt. These Governorates belong to the northern, southern, coastal, highland and desert regions of the country and include both production regions, as well as those were qat cannot be cultivated. The survey covered the main urban and several rural districts of each Governorate, and included areas where qat chewing had been practiced for hundreds of years (e.g. Ta'izz) as well as those where its use is of more recent origin (e.g. Hadhramawt). A total of 41 out of 332 districts inthe country were covered inthe surveylo. 16. The Qat Consumption Survey covered issues of education, chewing habits, expenditures on qat, respondent perceptions about chewing and its effects, and possible alternatives to chewing. The questionnaire was field tested in Sana'a City in April 2006, and after suitable revision, was launchednationwide in April and May of 2006. See Annex 1 for more detailsabout the characteristicsof the survey. A. Qat Consumptionis Widespread 17. Qat chewing is widespread amongst men and is on the increase amongst women. An average of 72% of Yemeni males reported that they chew qat, compared to 33% of females (Table 1). The actual proportion of females who chew qat may be higher still, considering that traditional cultural stereotypes discourage women from chewing qat. The survey results confirm that qat consumption is no longer confined to the elite, or dominated by men. Table 1: Sample PopulationChewingqat by Gender and Governorate Sana'a Capital Secretariat 78.0 53.2 Al-Hodayda 90.0 33.0 Ta'izz 86.1 55.3 Marib 71.3 32.0 Aden 83.2 19.4 Hadhramawt(Wadi) 27.7 4.4 Hadhramawt(Coast) 61.4 1.1 TOTAL 72.0 32.6 loThe HouseholdBudget Survey of 2005 also covered certainaspects o f qat consumption. Preliminary findings ofthe 2005 HouseholdBudget Survey are usedto augment those of the Qat Consumptionsurvey reportedinthis paper. 5 18. Yemeni men tend to under-estimate the prevalence of qat chewing amongst women. Husbandsreportedthat just 14%of their wives chew qat, while 33% of the women surveyed reportedthat they do indeed chew. Various reasons for this difference can be hypothesized. Insome cases, husbandsmay be unaware of the qat chewinghabitsof their wives. Inothers, they may simply be hesitant to acknowledge this because of the stigma attached to women's use o f qat, particularlyinthe more tribal parts of the country. 19. Men display addictive chewing behaviour. Some 42% of those men who use qat reportthat they chew every day of the week (figure 1). Ofthe rest, some 12%chew qat some 3-4 days per week, and the balance chew on 1-2 days of the week or less frequently. Thus, 54% of the Yemeni male populationof 12years of age and above chew qat more than three times aweek, and couldbe characterizedas displayingaddictivechewingbehavior. Figure 1: ChewingFrequencyof YemeniMales (in Percent,n=2,205: 50 40 30 20 10 0 Source: World Bank Qat ConsumptionSurvey, April-May 2006, (Gatter 2006) 20. Women users are also addicted; with one-fifth of all of the women in Yemen consume qat more than 3-4 times per week. While far less women than menchew qat, the proportionof those who do chew and are habitualin their use o f qat i s about the same. Some 13% of the women surveyed reported that they chew qat on most days of the week. An additional 7% reportedthat they chew qat some 3-4 times per week, 4% reportedthat they chew qatjust 1to 2 days of the week., and 7% reportthat they chew qat less frequently than 1-2days perweek (figure 2). Figure2: ChewingFrequencyof Yemeni Females(in Percent,n=1,774) Source: World Bank Qat Consumption Survey, April-May 2006, (Gatter 2006). 6 21. Pronounced geographic differences in qat use. Areas where qat i s cultivated and those where the habit has developed over a long period of time are more likely to report a higherconsumption incidence. The survey results show that the incidence of qat use is much higher in qat cultivation areas ofthe westernhighlands (Sana'a City and Governorate, Taizz) and in areas where consumption was introduced long ago (Le. al-Hodayda, Aden), than in areas where qat use is relatively new, such as Marib and Hadhramawt (Table 1). 22. Inthe traditional qat chewing areas, the vast majority of menreport that they regularly chew qat (Figure 3). For example, male qat use ranges from 78% in Sana'a Capital Secretariat to 90% in al-Hodayda. Inthe areas in which qat use is of more recent origin, such as Marib, male chewing incidence i s lower, but has already reached 71%. The Hadhramawt Governorate i s the region where the prevalence of male qat use is the lowest, with 61% o f the meninthe coastal region, and 28% inthe interior reporting regular use. Figure 3: RegionalDifferencesin Men's Chewing (in Percent; n=2,205) I $ 5% @Q* .*8`.a5& Source: World Bank Qat ConsumptionSurvey, April-May 2006, (Gatter 2006). 23. Geographical differences are more pronounced interms of women's use of qat (Figure 4). Consumption i s noticeably less amongst women in the interior tribal regions. In coastal Hadhramawt only 1.1% of women chew qat while in Ta'izz Governorate as many as 55% are qat consumers. Inthe highlands chewing incidencewas highest, ranging from 53% of women chewing in Sana'a to 55% in Ta'izz. Chewing among women in the coastal and desert areas of al-Hodayda, Aden and Marib i s much lower and ranges form 19% in Aden to 33% in al- Hodayda. Inthe interior of Hadhramawt Governorate, less than 5% of the women report that they chew qat. 7 Figure 4: RegionalDifferencesin Men's Chewing (in Percent; n=2,205) 60 40 20 0 Source: World Bank Qat ConsumptionSurvey, April-May 2006, (Gatter 2006). 24. Yemenis start in their teenage years. On average, the qat habit starts during adolescence. Where qat consumption is more widespread, and where the habit has developed over a longer period of time, children start consuming it at an earlier age (Table 2). For example, the average age at which consumption started was 16 years in Ta'izz, followed by an average of 17.2 years in Sana'a Governorate, and 17.6 years in the Sana'a Capital Secretariat. The age of first regular qat use was slightly higher in the areas where it is only produced in the more remote highlands (i.e. 17.8 years in Marib and 18.9 years in al- Hodayda). Distressingly, the survey found that children start chewing qat as early as age 5 in Sana'a, and at age 7 in Sana'a city and Aden. The highest starting ages were found in the former south, where qat consumptionwas either limited by the government to weekends (i.e. 20.9 years in Aden) or where it was completely illegal untilreunification ( i.e. 21.5 years in Wadi Hadhramawt and 23.6 years incoastalHadhramawt). Sana'a Gov. 17.2 Sana'aCauital Secretariat 17 h Ta'izz Marib Aden 20.9 21.5 Source: World Bank Qat ConsumptionSurvey, April-May 2006, (Gatter 2006). 25. Men chew steadily until old age, while women chew more as they get older. Men start chewing qat at an early age and the proportions of male chewers is relatively stable until the menreachbetween50 to 60 years of age. Incontrast to women, the proportion of elderly menwho chew qat drops markedly. While16% of menof 51-60 years of age chew qat some 3-4 times per week, only 3% of the men above age 60 chew qat. And while 41% of heavy chewers (i.e. those who use the plant 5-7 timedweek) of 51-60 years age chew, just 26% of the menabove age 60 reportthat they are regular chewers (Annex 2). 8 26. The older the women in Yemen, the more likely they are to chew qat. There is evidence that women tend not to chew before they are married, and thereafter the proportions of those who chew increaseas women grow older. The vast majority of the women inthe age category of 12 to 20 years in the survey sample were unmarried, and the proportion of those who were found to chew was just 18%. The incidenceof chewing then increasedsteady over the differentage groups ofwomen surveyed, with some 55% over the age of 60 reportingthat they do chew qat. Just 3% of the women below 20 years of age chew qat more than 3 times per week, compared to 11% of women between21-30 years of age, 18% of women between 31-40 years of age, and 29% o f women above 60 years of age (Annex 2). It appears that social pressuresare successful indiscouragingyoung women from using qat. These pressures tend to ease once women are married, and this is reflected in a tendency for women's consumptionto rise as they get older. 27. Home is where they chew. Nearly half of Yemeni males (49%) and females (44%) chew qat at home (Table 3). Some 21% of the men and 7% of the women chew mostly with friends away from home. A further 13% of males and 4% of females chew mostly at weddings or funerals. Source: World Bank Qat ConsumptionSurvey, April-May 2006, (Gatter 2006). 28. Women work and chew, while men relax and chew. Most of the men chew qat during time that is considered leisure, while just under half of the women combine work, in the householdor on the farm, with qat chewing. Overall, some 75% of qat users declare that they chew during their leisure hours. Of this 44% of the women report that they chew qat while working, comparedtojust 17%of the men. B. Reasonsfor Chewing and Abstaining 29. Why start? Most consumers report that they start their qat habit because of peer pressure (Table 4). Chewing is a very social habit and practically all Yemeni afternoon gatherings take place around a qat chewing session. Some 66% of the men and 72% of the women claim that the reason that they started to chew qat is because "all myfiiends and family chew". Lack of alternative leisure activities also plays a role in the decision to start using qat. Some 19% of the men and 10% of the women claim that they started to use qat because "there was nothing else to do." A number of men state that they started chewing because it enables them to work (8.6%) or study (3.4%) more effectively. Some 9% of the women reportedthat they startedto use qat becausetheir husbands asked them to. 9 There is nothing else to do 19.0 10.0 To be able to work 8.6 2.3 Inorderto study 3.4 2.6 Becausemy husbandwants me to 0 9.0 To alter my mood 0.9 0.7 To attendweddings or funerals 0.8 0.2 Inorderto forget 0.5 0.9 Others 1.3 1.9 Total 100 100 30. Why use qat. Men and women have similar reasons for using qat (Table 5). The most frequently observed reasons are that "it gives me strength" (12% o f male and 14% o f female respondents), "it makes me more alert" (12% o f men, 12% o f women), "it helps me concentrate" (14% o f men, 10% o f women), because my friends chew (9% o f males and 8% o f women), and simply because "Ilike it" (9% o f men, and 11% o f women). Qat i s also used, by some, as an escape from the pressures o f daily life. Some 6% o f men and 7% o f women namedrelief from depression as a reason for chewing, and another 5% o f men and 7% on women claimed that it helps them toforget. Table 5: Reasons Why Consumers Use Qat Source: World Bank Qat ConsumptionSurvey, April-May 2006, (Gatter 2006). 31. Users clearly do not use qat for its health benefits. Indeed, just 2% o f the male respondents and 5% o f the female respondents reported that they use qat because it i s perceivedto be good for their health. Moreover, and in stark contrast to popular folklore, just 2% o fthe male respondents reportedthat qat use is an aphrodisiac or performanceenhancer. 10 32. Why some abstain? Most of Yemen's non-chewers claim that they don't use qat for health (35% of men, 33% of women) (Table 6). The second most important reason for abstaining i s the highcost of usingqat. Some 27% o f male non-users and 19% of female non- users claim that they don't use qat because it would be a waste o f money to do so. Religious reasons also play an important role inthe decisionto abstain from qat use. Nearly a quarter of the respondents claim that they don't use qat because the Qur'an would prohibit it, either as an intoxicant, or because spending on it could be considered a sinful waste and an extravagance. Interestingly,some 23% o f the women don't use qat becausethey don't like the taste of its bitter leaves, while just 11% o f the male non-users cite taste as a reason for abstaining. Table 6: ReasonsWhy Consumers Abstain from Qat Reason Male Female Healthreasons 35.2 33.4 Financial reason 27.2 18.5 Religious or legal-religious (Sharia)reasons 24.4 23.2 Don't like taste 11.0 23.4 Others 2.1 1.5 Total 100.0 100.0 C. Qat Use Adversely Affects Health and Family Life 33. The epidemiology of qat use has been subject to extensive study. While qat i s still used in traditional Yemeni medicine to combat melancholy, to treat cold and diabetes, its adverse effects on human health are well-known. These include sleeplessness, constipation, gum disease, teeth decay, and pesticide poisoning. 34. Adverse health-effects of qat use are widespread. The results from this survey confirm that the adverse effects o f qat consumption are widespread----some 80% of all male users and 70% of all female users report health disorders that they perceive are related to qat consumption''. Some 29% of male and 23% of female users are reported to suffer from sleeplessness, with its negative implications for day-after work performance, social relations and psychologicalwell being. I', The preliminary findings of the 2005/2006 Household Survey confirm the seriousness of the adverse Qat health effects. According to the Household Survey, Qat users spent an average o f 886 Riyal for the three months prior to the survey on medicinesand healthtreatments comparedtojust 440 Riyal for non-users. 11 Table 7: Consumer PerceptionsRegardingthe Health Effectsofqat Effects on your heafth Male Female INO effect I 20.0 I 30.2 I Sleeplessness 29.1 23.0 Constipation 18.4 15.0 Other Gastrointestinaldisorders 14.0 14.5 Oral disorders 12.7 9.5 Highblood pressure 5.6 7.4 Other Illnesses 0.2 0.3 ITotal I 100.0 I 100.0 I Source: World Bank Qat ConsumptionSurvey, April-May 2006, (Gatter 2006). 35. A large share of the Yemeni population is reportedto suffer from hemorrhoids, which is closely linked to qat consumption. In the survey, some 18% o f male users and 15% o f female users were found to be affected by constipation. Other gastrointestinal disorders were reported by a further 14% o f male and 15% o f female qat users. An additional 6-7% of the respondents felt that qat was a cause o f their highblood pressure. 36. Qat chewing can have adverse effects on the gums o f chewers since it i s chewed for hours and the leaves have a high content o f acidic tannins. Pesticide residues aggravate this problem and a number o f cases o f oral cancer have been associated with qat chewing. Inthis survey, some 13% o f male qat users and 10%o f female qat users were reported to suffer from a variety o f oral disorders. 37. Consumers self-treat qat-related healthdisorders. Over 80% o f all chewers report that they treat themselves by different traditional and "modern" means to combat the unwanted side effects o f qat use (Table 8). An overwhelming number o f respondents drink milk (43% o f male and 33% o f female users) to combat sleeplessness, a traditional remedy used for many generations in Yemen. Some 26% o f male and 24% o f female users force themselves to eat dinner, despite the loss o f appetite resulting from qat use. Eating after qat use helps to overcome the effects o f the acidic tannins that can cause gastrointestinal disorders and constipation. Between 5-7% of all users consume sweets and/or fruits to boost their blood sugar level, while approximately the same proportion take medicines to overcome various adverse health side effects arising from qat consumption. 12 Table 8: Ways ConsumersTreat Adverse Health Effectsof Qat Treatment Male Female Donot treat myself 10.5 19.7 Drinkmilk 43.3 33.4 Have dinner 26.2 24.0 Eat sweets 6.9 6.2 Eat fruit 5.3 6.6 Takemedication 5.2 7.1 Others 2.7 3.0 38. Household stress. Many Yemenis claim that qat consumption is one of the main reasons for family breakdowns, since continued absence of one or both of the partners from home during the afternoon and evening hours, as well high qat expenditures causes severe stress on the families. Indeed, some 53% of male and 60% of female chewers believe that qat chewing is a source of friction within the family. Moreover, as many as 29% of women respondentsreportedthat qat is "sometimes" a cause of conflict with their spouse and a further 8% reported that conflicts "often" arise due to qat use. Men, on the other hand, believe that qat consumption does not necessarily lead to family conflicts. Some 18% of the male respondents were reported to "sometimes" have qat-related family conflicts, and just 6% reportedthat they often haddisputes with their wives over qat. D. Qat UseDrainsProductivityandFuelsPoverty 39. Qat use is extremely time consuming. Qat users spend close to one-third of their waking hourschewing (Table 3 and Figure 5). Muchofthis is time that workers inother low- middle income countries would spend engaged inproductive activities. Only 7 percent of qat users spend less than 2 hoursa day chewing. Some 36% ofthe users spend 2-4 hoursper day; 35% spend 4-6 hours a day, and a whooping 22% spend more than 6 hours a day chewing qat. Inurbanareas, afternoonsare effectively lost to qat chewing, particularly inthe public service and in the services sector. The fact that only 24% of chewers work while chewing, whereas the overwhelming majority of Yemenis (75%) chew during their leisure time and are either at home or with friends, is added evidence of a tremendous national productivity loss (see Table 3 and Figure 5). 13 Figure 5: Amount of Time Spent ChewingQat Per Day (in Percentof total Chewers) I 40 I 35 I 30 25 20 15 10 5 0 less than 2 hours 2 4 hours 4-6 hours more than 6 hours Source: World Bank Qat ConsumptionSurvey, April-May 2006, (Gatter 2006). 40. The amount of time devoted to qat chewingalso varies by region. Inal-Hodayda, for example, just 13% of the respondents chew for more than 6 hours per day. In Aden, by contrast, more than 33% of the respondents chew for 6 hours or more per day. In regions where the qat habit is of more recent origin, the evidence suggeststhat those who chew do so for a longerperiodoftime each day (Table 9). Aden 2.9 22.9 41.O 33.3 Hadhramawt(Wadi) 2.7 36.6 42.9 17.9 Hadhramawt(Coast) 4.5 23.1 45.5 26.9 Average 6.9 36.0 34.8 22.4 41. Qat has mixed effects on work performance. There is anecdotal evidence that qat chewing may lead to better same-day work performance, but that it also leads to a loss of sleep and thus to reducedwork performance the next day. This is, in fact, what the majority of Yemeni chewers believe. Some 78% of the male chewers and 69% of the female chewers believe that qat chewingboosts their immediate work performance (Table 10). Only 6% of the male and female chewers believethat ithinderswork performance, while 16%ofthe male chewers and 25% of the female chewers report that it has no effect on work performance. While many Yemenis reportedthat there was no effect of qat on their work performance the next day (60% o f female and 43% of male users), nearlyone third of all male users and more than a fifth of all female users report that they experienced negative effects on their work performance from chewing qat the previous day (Table 11). Hence, for many qat users, the day-after effects o f sluggishness, sleepiness, and an inability to concentrate off-set the more immediateenergy boostthat qat consumption brings. 14 Table 10: Effect of Qat on ImmediateWork Performance (in Percent:n=2.275). Effect Male Female I Better I 78.1 I 69.3 I Worse 6.2 5.7 No Effect 15.8 25.0 Table 11: Effectof Qat on Day-After Work Performance (in Percent:n=2.297) Positive 25.2 18.4 Negative 31.7 21,5 No Effect 43.1 60.1 42. In addition to diverting time that could be devoted to productive activities, qat also contributes to poverty by pre-empting a substantial share of the household budget. National household expenditure surveys (Table 2, Annex 3) show that 9 to 10 percent of household expenditures are devoted to buying qat throughout all income classes. Amongst the poor, outlays on qat are greater than, for example, private expenditureson healthor education. 43. During the survey period, the daily pay of day laborers averaged YR400 per day, while the daily pay of civil servants was closer to YRlOOO per day. The cost of a daily portion of qat ranged from YR200 to W O O 0 depending on quality and location, with an average price of about yR450 per day12. For the most habitual users, it is not uncommon for half or more of the daily wage to be devotedto buying qat. 44. Yemeni consumers are aware that qat consumption crowds-out other essential householdexpenditures. Some 94% of non-chewers and 77% of chewers report that qat has a negative effect on the family budget. Close to two-thirds of all of the women, and about half of the menbelievethat they are unable to pay for qat and meetthe other essentialneeds of the householdat the same time. Inthe poorer regions, the trade-offs betweenqat consumption and the resources required to meet basic needs is far more pronounced. Incoastal Hadhramawt, for example, some 87% of the women believe that families will be unable to afford qat and meet other essential family requirements. In Wadi Hadhramawt, 77% of the men share this view. Gatter, Peer (2006 forthcoming); A Survey o f Sana'a Qat Markets during 2000 and 2005. 15 Table 12: PopulationWho Cannot Afford Qat and Meet Their Basic Needs Sana'aGov. 34.5 47.6 Sana'a Cauital Secretariat 44.3 48.5 al-Hodayda 32.5 58.9 Ta'izz 32,s 58,9 Marib 48.1 55.8 Aden 39.6 54.3 Hadhramawt (Wadi) 76.7 80.2 Hadhramawt (Coast) 51.4 86.5 45. The Qat debt trap. For some families, the qat habit leads down the road to a never- ending cycle o f indebtedness. Close to a fifth o f all families are forced to borrow simply to finance their qat habit. Some 18% o f all male qat users and 17% o f all female qat users reported that their families were forced to borrow to finance regular qat purchases. 46. 2. Qat crowds out spending on basic needs. What are the opportunity costs, within the household, of the money that is spent to buy qat? To answer this question, respondents were asked to assess what it is that they would purchase if extra funds were available (Table 13). One-fifth o f all o f the respondents replied that they would spend additional funds on food, and another fifth gave priority to basic medicines; another 15% reported that they would spend the extra money on consumer goods; while 14% would use additional funds to augment savings. Another 14% of the respondents would devote additional resources to their children's education; and another 11% would use the funds to help relatives in need or to support the poor. Just 4% o f the respondents said that they would use additional monies to buy more qat. 47. Qat and governance. Qat also contributes to poverty by fostering what the population perceives to be a climate conducive to corruption. Empirical evidence based on cross-country comparisons shows that corruption has large, adverse effects on private investmentand economic growth, and that more corrupt governments are less likely to spend on education and poverty red~ction'~. i s particularly revealing that some 95% o f all of the It survey respondents believe that qat chewing leads to corruption in Yemen, in one form o f another. l3Paolo Mauro, 1996, "The Effects o f Corruption on Growth, Investment, and Government Expenditure," IMF Working Paper 9619. WashingtonD.C. 16 Table 13: What would respondents purchase ifthey had more funds? E. Consumers Need Government Help To Kick the Habit 48. Qat chewingis a bad habit. personalConsumersare well aware that qat has adverse and social consequences (Table 14). They believe it is generally a bad habit. Some 96% of non chewingmales and 92% o f non-chewingfemales describe qat chewingas a "bad habit". What is especially striking i s that some 72% of Yemeni men who chew qat and 69% of the chewingwomen share the same opinion. While most of the consumers don't believethat qat chewing is against religious law, one-fifth of the Yemeni chewers do believe that they are committinga sinby chewingqat. Table 14: Consumer Perceptionsof Qat Chewing 49. Qat harms the economy. Most Yemeni's believe that qat use has adverse effects on the economy. Some 75% ofthe users and 94% ofthe non-usersreportthat qat places a heavy strain on the family budget. A similarly high proportion feels that qat use has adverse economic impacts and that its production and trade creates opportunitiesfor corruption. A substantial share of the respondents (86% of users and 98% of non-users) feel that qat is an importantcause ofpoverty, andmoreover,that its use is badfor Yemen's nationalimage. 17 50. Qat is believed to harm health and family life. The vast majority o f Yemeni's believe that qat use has negative health effects (i.e. 78% o f users, and 95% o f non users) and contributes to the erosion o f positive social values (i.e. 66% o f users and 86% o f non users). Most believe that destroys social values and moral (64% o f users and 93.0% of non-users). Just over half o f all users believe that qat contributes to tensions within the family, while between 77-90% o f non-users believe this to be the case. Users are equally divided on the question o f whether or not qat contributes to domestic violence, neglect of children and divorce, while non-users overwhelming believe that it does so. 51. Help wanted. If consumers know that qat is bad for them, and moreover bad for society as a whole, why don't they just stop? The answer to this i s simple---they can't. Either because o f social pressures, or because o f the psychological dependency resulting from prolonged use, consumers do not feel that they can stop usingqat on their own. Some 53% of all male and 61% of all female respondents declare that Government intervention is necessary to address the qat problem (Table 15). Amongst those who support Government intervention, the vast majority stated that they would welcome some form o f legal restriction on qat use. Only 37% o f the men and 24% o f the women didn't want any government intervention at all, and about 10-15% o f the respondents had no view on this matter. Table 15: Should the GovernmentLimit Qat Chewing? (in Percent: n=3.889) No 36.5 24.1 Idon'tknow 10.3 14.7 52. Respondents were also asked to identify feasible alternatives to qat consumption (Table 16). A large number o f the respondents were unable to identify any suitable alternatives, itself a sign o f the unique role that qat consumption plays in everyday life. Of those who did respond, about half felt that the creation of greater economic opportunities, either in paid employment or in cultivating qat substitutes, could be a far better use o f their time. About one-quarter o f those who did respond to this question felt that the Government could help by investing more in alternative leisure time activities. It i s also important to note that very few felt that an outright ban on qat consumption would be a feasible option. 18 Table 16: Respondent's Suggestions for Possible Alternatives to Qat Go to mosque or readthe Qur'an I 75 1.7 II Listen Read or studv 62 1.4 to lectures 70 1.6 I Go to an internet cafe II 9 0.2 Consumption Eat more, especially fruit 33 0.8 Alternatives Use alcohol or other drugs 8 0.2 Chew gum 6 0.1 Drinktea I -...... .. 12 0.3 ISubtotal I I .- 59 1 1.3 I Idon'tknow I 78 1.8 19 111. DiscouragingQat Consumption A. ImplementingPastPolicyDecisions 53. Responding to NGO initiatives, the Government has begun to take actions to curb qat consumption. Qat is the only agricultural commodity whose sale is subject to taxation, and its production is discouraged by exclusion o f qat farmers from agricultural services, includingresearch, extension, and public credit and irrigationimprovementprojects. Qat is prohibitedfrom use in government offices; police and soldiers are forbiddento use qat while on duty; and some qat markets have been moved outside of the city centres. The Government has introduced the five day week and longer working hours, both aimed specifically at discouragingqat use. The Government of Yemen also carried out five years ago an extensive review of the qat situation, the findings of which were discussed in a national conference in April 2002. Stemming from this conference, a series of new decrees were passed, including policy measures aimed at discouragingharmful pesticide use in qat production, promoting alternative crops, and enhancing access to alternative recreational facilities14. 54. Implementationof the post 2002 reforms has provendifficult, particularly in the face of stiff resistance from influentialqat traders andproducers. As ofmid-2006, qat is once again being chewed openly by the military in uniform and in public buildings from which it was bannedin 1999. Effortsto relocate qat marketsto the outskirto f the main cities have stalled, and qat chewingsessions appearto start earlier each day andto continuefor muchof the day. 55. The still widespread use o f qat, as the results of this survey attest, imply that stronger policymeasureswill be neededto curb its use. Public policiesaimedat combatingqat must, however, contend with the fact that it is an integral part of Yemen's culturalendowment, that it plays a vital role in rural economies, and that the income flows from qat production are a key source of local government revenues. A further problem with qat is the way in which control over its productionand distribution is exercised. Those who grow qat are so powerful that there is a highpolitical risk to efforts madeto campaignagainst it. Moreover, those with a vested interest in the qat status-quo have proven to be extraordinarily effective in circumventingpoliciesaimed at curbingits use. 56. The Government is fully aware of just how difficult it will be to combat qat. Accordingto the Third FiveYear SocioEconomicPlanfor PovertyReduction(DPPR): "the Government intends to address the qat phenomenon in an objective, informed and gradual fashion, while striking a balance between its economic, social, health and environmental aspects (p.6)." Emphasis in the DPPR is accorded to water conservation and to the introductionof new cash crops, both of which could be ex ected, with the passage of time, to encouragefarmers to shift resources out of qat production . IP l4This included a decree on pesticide use (Cabinet Decree No. 135 of 2002) aimed at improving consumer safety. It calls for (a) inspection campaigns on chemicals in all governorates; (b) confiscating and destroying smuggled chemicals; (c) checking legally imported chemicals for validity and conformity with the established specifications;(d) maintaining a recordof registeredpesticideimportersand conducting spot checks on them. l5Ministry of Planning and International Cooperation. 2006. Yemen's Third Socio-Economic Development Planfor PovertyReduction(2006-2010). 20 57. While the general trend has been for a rapid increase in qat cultivation, some success has beenregistered incrop substitution,albeit on a small scale. Since 1999,the IsmaeliBohra communityinthe Haraz regionhas successfully replaced qat by other economic activitiesand has uprooted over 200,000 qat trees. Alternative income generation activities include the cultivation of coffee, almonds, mangos, as well as goat herding, beekeeping, embroideryand sewing. There has been also investment in rural transport systems. An assessment of the strengths and weaknesses of the Bohra project could be used as the basis for a more comprehensiveqat substitutionprogram16. 58. It may, however, be difficult to dislodge qat from highlandcropping systems. Efforts to promote water conservationtechniques inthe World Bank assisted Sana Basin projecthave resulted infarmers expanding the area allocatedto qat cultivation. Similarly, efforts aimed at phasing out diesel subsidies have resultedin farmers puttingmore, rather than less, emphasis on qat cultivation, becauseof its superior profitability. 59. Over the past three decades, Yemeni farmers have shownthemselves to be responsive to market forces when it comes to decision-making. Lands have been diverted to qat production because of growing market demand, a well-organized distribution system, and above all, the fact that profitsto qat productionandreturns to laborwere far superior to that of any other competingcash crop. Encouragingfarmers to divert resourcesfrom qat cultivation may be difficult, particularly given the high profitsthat the crop attracts. Inthe long-run, an importantway of reducingthe profitability ofthe crop would be to adopt economic and social policiesthat trigger a reductioninconsumer demand. B. Directions for Future Policy Dialogue Lessons from International Experience 60. International experience with drug demand reductionprovides some lessons that can usefully be applied in the Yemen context, given recent interest within Government to reinitiate efforts to address this major development challenge. The Political Declaration adopted by the UN member states at the General Assembly Special Session on the World DrugProblemin 1998 recognises that drug demand reductionis indispensable in solving the drug problem. It identifies a set of agreed principles for Government's to adopt in their efforts to support the reduction in drug use demand (Box 2). Many low income countries have incorporateddrug demandreductionelements into their nationaldrug control strategies. Experience to date with drug demand reductionefforts in low income countries suggeststhat both awareness building and risk reduction is needed to effectively reduce demand. In all cases, communities must play an active role in defining the drug problem and finding a suitable solution. Experience also shows that different approaches are required for different groups inaddressingdifferentneedsand situations17. l6See Gatter (2005) for a preliminaryreviewofthe IsmaeliBohra community gat substitution program. United Nations Office for Drug Control and Crime Prevention. 2002. Lessons Learned in Drug Abuse Prevention, A GlobalReview, New York. 21 Box 2: Principles of Drug DemandReduction The UNDeclaration o fGuidingPrinciples o fDrugDemandReduction(1998) include: An integrated approach to solving the drugproblem should be adopted and should consist o f a balance between drug demand reduction and supply reduction Demandreduction policies shall: -- aim at preventing the use o f drugs and adverse consequences o f drug abuse provide for and encourage active and coordinated participation o f individuals at the -- community level be sensitive to both culture and gender contribute towards developing and sustaining supportive environments Demand reduction strategies should be based on a regular assessment o f the nature and magnitude o f drug abuse and drug-related problems inthe population Demand reduction programmes should cover all areas o f prevention, from discouraging initial use to reducing the negative health and social consequences o f drug abuse A community-wide participatory and partnership approach is crucial to the accurate assessment o f the problem, the identification o f viable solutions and the formulation and implementation o f appropriate policies and programmes Demand reduction efforts should be integrated into broader social welfare and health promotion policies and preventive education programmes Demand reduction programmes should be designed to address the needs o f the population in general, as well as those o f specific population groups, with special attention being paid to youth Information utilised in educational and prevention programmes should be clear, scientifically accurate andreliable, culturally valid, timely, and tested with a target population States should place appropriate emphasis on training policy makers, programme planners and practitioners inall aspects o fthe design, execution and evaluation o fdemand reduction programmes and strategies Demand reduction strategies and specific activities should be thoroughly evaluated to assess andimprovetheir effectiveness. The results o fthese evaluations should be shared. Source: Special Session of the General Assembly Devoted to Countering the World Drug Problem Together, 8- I O June 1998: Political Declaration; GuidingPrinciples of Drug Demand Reduction; and Measures to Enhance International Cooperation to Counter the WorldDrug Problem. 61. A key lesson from international experience is that once there is a strong negative stigma attached to drug use (and abuse), new users will be discouraged from starting, and demand amongst already established users will fall. Experience worldwide shows that efforts to address drug reduction require long time-frames, and persistent support from national leadership. But while the benefits take time to accrue, international experience provides unambiguous evidence that deeply engrained attitudes can be changed. For example, in the 1950s, tobacco consumption was viewed as glamorous and prestigious in most o f the high- income economies. After five decades of concerted demand-reduction policies, cigarette consumption inthe OECD nations is widely discouraged (Box 3). 22 Box 3: OECD Tobacco Consumption: FromGlamour to SocialStigma Anti-smoking groups, along with many government health departments have attempted to counter the advertising of tobacco by creating their own advertisementsto highlightthe negative effects o f smoking. The earliest commercials mainly focused on the increased risk o f lung cancer and the problems associated with passive smoking. However, they have become increasingly hard-hitting and more targeted to youth over the years, with some campaigns now centred around decreased physical attractiveness. Public finding has played an important role in public campaigns to discourage smoking. The Britishgovernment spent E31 million in2003 as part of their anti-smoking campaign. In2005 the European Union launched the "For a life without tobacco" campaign in all its constituent countries to helppeople quit smoking. Laws implementing bans on smoking have been introduced by many countries invarious forms over the years, to complement cigarette taxes and public awareness campaigns. Italy banned smoking from the 10th January 2005 in all enclosed places except private homes and specifically designated areas satisfying strict technical conditions. Despite initial skepticism, the measurewas well accepted and is generally respected. France will ban smoking from February 2007 in all public places, and opinion polls suggest that 70% of people support the ban. Scotland prohibited smoking in all enclosed public places on March 26, 2006, after the Smoking, Health and Social Care Act took effect. England and Northern Ireland have both prohibited smoking in all enclosed public places starting in2007. Experiments with outdoor smoking bans in specific contexts, especially in public or government- owned spaces, have begun. California bans smoking in all buildings open to the public. Additionally, California prohibits smoking within20 feet o f any door or window o f any government buildingwithin the state. Smoking has been bannedinTokyo's Chiyoda Ward since October 2002. In addition, there are some 60 other municipalities in Japan which have regulations to ban or discouragepeople from smoking on the street. Source TobaccoAdverstising. 2006. http://en.wikipedia.org/wiki/Tobacco-advertising, and Smoking Ban 2006. http://en.wikipedia.org/wiki/Smoking-ban Elements of a Qat DemandReductionAgenda 62. Securing the Support of National Leadership. Past experience suggests there is a needto engagethe top leaders, at all times, ina campaign to discourage qat consumption. The office of the Presidency, the Parliament, and local government leaders all must be informed and involved in campaigns to discourage qat use. Ensuringthat national leaders have access to the findings of this survey, and are aware that the majority o f consumers want Government's assistanceto help them reduce qat consumption, would be a good start towards buildinga solid base o f leadership support. Bearing inmind, however, that qat is regarded by many as a unique element o f Yemeni culture, international agencies should be seen as supportive o f local initiatives to curb consumption while beingrespectful o f Yemen's unique cultural endowments. 63. Enforcing the post-1999 Reforms. Renewed efforts to enforce policy reforms launched since 1999 to reduce qat consumption would send a signal to the population that 23 Government leaders are aware of the need to take steps to curb excessive qat consumption. This would include enhanced enforcement of public policies aimed at prohibiting qat consumption in public buildings and by serving military officers; ensuring that government office hours last until3 pm inthe afternoon; and actively relocating qat markets out o f the city centers. Inparticular, a policy o f zero-tolerance for the use o f qat during working hours, and leaving the place of work before the end of statutory work hours, both for public servants, the army and the policy, would send a powerful signal to the population-at-large that qat consumption will no longer be allowed to undermine the effective operation o f the public service. 64. Increasingthe Tax Wedge. One way o f reducing qat demand i s to simply make qat more costly to consume. As the price o f a commodity rises, the quantity demanded shall fall. Raising taxes on qat can serve to reduce demand, and therefore also lower margins to traders and producers. Although it i s widely presumed that qat consumption i s price inelastic, the results o f various socio-economic surveys suggest that consumption is fairly elastic up until the highestincome brackets. 65. Qat remains lightly taxed, because the nominal rate i s low, and enforcement is plagued by avoidance, fraud and ineffective tax administration. Although Qat is taxed at a nominal 20% ad valorum rate, the effective tax rate is much lower---at just 3 to 4% of the value of final consumption. Boosting the tax rate, and substantially improvingtax enforcement, could help serve to deter excessive consumption, particularly amongst the poorer segments of society. In addition, five percent of what i s actually collected is, by law, contributed to the Government's Fund for Youth and Sports. Increasing the tax take could therefore automatically boost resources available for alternative leisure activities. 66. Worldwide, tobacco is taxed at a far higher rate than qat is taxed at in Yemen in order to discourage consumption. In high-income nations, taxes amount to two-thirds or more of the retail price of a pack of cigarettes. Inthe lower-middle income countries, by comparison, taxes amount to about half of the retail price of a pack o f cigarettes'*. This suggests that taxes on qat use could rise substantially inorder to discourage use. 67. In designing an effective qat tax system, the government needs to take into account several factors, including the possible impact of smuggling, the purchasing power o f the local consumers, rates in neighbouring countries, and, above all, the ability and willingness of the tax authority to enforce compliance. Since the primary purpose o f the tax is to discourage qat consumption, a strong case can be made for specific taxes (Le. based on quantity) that would impose the same tax per bag sold instead of an ad valorum tax. Specific taxes are easier to administer than ad valorum taxes because it i s only necessary to determine the physical quantity o f the product taxed, and it i s not necessary to determine its value. The use of tax- stamps can also assist inensuringthat qat taxes have been paid, and that that goods that have paid the tax appropriate for onejurisdiction are not shipped to another. 68. To be effective, any government that decides to implement a vigilant qat control policy must do so ina context inwhichthe decision has broad popular support. While it might seem that qat consumers would be strongly opposed to qat control, the reality is strikingly different: the findings o f this survey show that qat consumers want the government to World Bank. 1999. Curing the Epidemic: Governmentsand the Economics of Tobacco Control. Washington D.C. http:/lwww 1.worldbank,ore/tobacco/reports.htm 24 intervene to curb consumption. They are especially concerned that their children not take-up the habit, and they want the Government to help foster a social setting in which qat consumptionis stigmatizedand alternatives to its use are promoted. 69. BuildingPublic Awareness. Another way that the Government could discourage qat consumption is to inform consumers about the adverse consequences arising from prolonged use. All marketeconomies suffer from information failure-4.e. consumers simply don't have all the facts. In the face of information failure, the Government plays an important role in helping consumers make an informed-choice about the products that they consume. At this juncture, there is a crying need for Government to provide Yemeni consumers with information about the hazards of qat consumption, both to themselves and to society as a whole, and to provide positive role models as advocates for change. A well-focused information campaign could build on the findings of this survey, as well as other sources of scientific information on the health, productivity loss and poverty impacts of qat consumption. Likewise, information on the adverse effects of qat consumption should be introduced in the education curriculum, and mainstreamed into health awareness programs. Endorsement by leading public figures of the merits of breaking the qat habit, peer to peer education, and involvement of the private sector in social marketing campaigns should all be considered as components of a multi-pronged strategy for building well-informed, public awareness and understanding. 70. PreventingYouth from Starting the Habit. Public awareness campaigns should be designed with the specific interests of different age groups in mind. The survey results suggest that qat consumption begins in the teenage years. The Government has an important opportunity to help prevent the qat-habit from forming by targeting information campaigns specifically to the youth population. In 2006, the Yemeni soccer team dropped out of the Asian Games in Qatar because they knew they would fail a drug test. Incidents such as this, together with the findings of the Qat Consumer Survey, could serve as the basis for an information campaign aimed at convincing the youth that qat consumption is not intheir best interest. Efforts should also be made to involve civil, religious and tribal leaders who appeal to youth in such outreach campaigns. Targeted efforts to prevent qat consumption could payoff rather quickly, in terms of reducing demand, since nearly half of the Yemeni population is below age 15. 71. There may also be opportunities to scale-up public awareness efforts piloted by civil society organizations. On a small scale, several NGOs have recently been active in fostering public awareness about the hazards of qat consumption. These include the Ta'izz based "Anti qat Association" (AQA) (Jami'aMuwajiha Adrar a1 Gat), the Sana'a based Afif Cultural Foundation (Mu`assasaal- Yfif al-Thaqafiya)and its associated National Society for Combating Qat Disadvantages(NSCQD) (al-Ja`miatal-Wataniyali MuwajihatAdrar al-qat), the Aden Society for Combating Qat (Jam`iatMakafahat al-qat),and the Ibb based Friends without Qat (Asdiqa'bi la qat).The Government could work inpartnershipwith these NGOs and build on the lessons learnedfrom their outreach efforts to scale-up a more wide-ranging public awareness campaign. 72. DevelopingAlternative RecreationalFacilities. One of the findings of the 2002 Qat Conferencewas that there was a pressing need for expanding access to recreational facilities for young people so that they would have a viable alternative to the ubiquitous qat party. The findings from this survey support that conclusion. In addition to peer pressure, one of the reasons that young people give for starting to use qat is that there is a perceived lack of 25 alternative recreation facilities. Improving access to sports facilities, youth centers and boosting public financial support for youth-oriented cultural activities could help to provide alternatives for the leisure time o fthe youth. 73. Stimulating Private Sector Developmentto Boost Employment. The greater the employment demand, the less likely consumers are to allocate their increasingly valuable time to qat consumption. Just under half o f the respondents identified the need to expand economic opportunities as a way o f encouraging a reduction in qat use (Table 16). Public policies aimed at boosting competitiveness, buildinghuman capital, and enhancing the private sector enabling environment can make an important contribution to reducing qat consumption by boosting the demand for labor, and therefore by raising the opportunity cost of the many hours that laborers devote to consuming qat. Certain policies aimed at boosting macroeconomic stability and competitiveness, such as the gradual reduction in diesel subsidies, will also free-up resources that can be used for financing measures aimed at reducing qat demand. Inaddition, a phased elimination of the diesel fuel subsidy would boost qat production costs since diesel fuel i s used to power the tubewell pumpsthat irrigate qatlg. 74. Supply-side Challenges. There i s a need to identify viable crop diversification programs, ifviable substitutes for qat inthe rural economy are to be created. Inthe highlands, many other crops could be cultivated, but there is little evidence that such substitutes could generate profits or employment equivalent to that which qat cultivation and marketing provide. Efforts to combat drug cultivation in other countries have tended to require a combination of laws restricting production, crop diversification programs, programs that ban harmful pesticides and tax other agro-chemicals, and safety-net measures. There i s a need to anticipate and mitigate the adverse economic effects o f what could become a secular decline indemand, giventhe importance of qat inbothemployment and inthe rural economy. More research i s necessary to identify an appropriate package of supply-side interventions-4.e. legal measures, agriculture, pesticide-control and water resource programs, and safety net interventions (such as the use o f Social Funds and Public Works Programs) that could be used to mitigate the employment and rural income effects o f declining qat demand. 75. AddressingKnowledgeGaps. While epidemiologicalresearch into the consequences o f qat consumption has a long history in Yemen, and much is known the short-term health effects of qat use, the biological and social impacts o f prolonged use are not well understood. In addition, with the recent upsurge in pesticide use on qat, more research is needed on the impact of pesticides inpregnant women, lactatingmothers, and their off-spring. 76. Druguse by a group of people is rarely causedby a single factor. Identification ofthe nature and type o f drug abuse in a certain target group should precede and inform any drug abuse prevention intervention. Compared to the numerous epidemiological studies, rresearch on the causes and consequences of qat consumption, and the behavioural pressures associated with qat uptake are limited in number and scope. While demand-control interventions are being implemented, parallel research activities into these issues may help to refine the targeting o f interventions, such as those designedto improve health information for the poor, for maximum effect. Evaluation should be integrated into qat reductionpolicies, projects and 19 Domesticdiesal prices remainfar below world market levels and total subsidies are estimatedat 10percent of GDP. The 2005 experience suggests the merits of boosting diesel prices in a series of steps, and with accompanyingpro-poor measures aimed at mitigating the adverse social impact of price increases. Diesel prices were raised in one step from YR 17/litre to YR 45/litre in 2005, which triggered demonstrations,riots, loss o f life and destruction of property. Inresponse,the governmentrolledbackthe price increaseto YR 35/litre. 26 programmes from the outset, and should continue throughout. This will help to establish evidence o f effectiveness and fuel learning processes intrinsic to the success o f the demand reductioneffort. 27 References Alles, G., M. Fairchild & M. Jensen (1961): Chemical pharmacology of Catha edulis. J. of Medicinal andPharmaceuticalChemistry,3: 323-352. Amer, J. & A. Juneid (1996): Economics of Qat Production. Paper for the National Policy Workshop on Agricultural Development. Sana'a, September 1996. Brenneisen, R., H. Fisch, U. Koelbing, S. Geisshusler & P. Kalix (1990): Amphetamine-like Effects in Humans of the Khat Alkaloid Cathinone. Brit. J. of Clinical. Pharmacology (Dec.) 3016: 825-828. Demisew, S. (1984): Botanical Aspects of "Khat" Catha edulis (Celastraceae ,p. 5-11. In: E. D a p e (ed.), Proceedings of the Int. Symposium on Khat (Catha edulis , Addis Abeba, 1 Ethiopia, Dec. 15, 1984. Gatter, P. (2005): Successes and Failures of the Qat-Substitution Programme o f the Ismaeli Bohra Community in the Haraz Mountains. An Analysis for the Ministry o f Water and Environment, Sana'a, 31pp. Gatter, P. (2006): Socio-Economic Assessment o f Qat Chewing inYemen. A Survey o f SevenGovernorates. Report for the World Bank, 62 pp. (Mimeo) Gatter, P. (2006bin print):A Survey o f Sana'a Qat Markets during2000 and 2005. Gatter, P., Q. Abdul Malik & K. Sae'ed (eds.) (2002): First National Conference on Qat. Under the Auspices of H.E.the President of Yemen, Mr.Ali Abdula Salleh. Proceedings and Conference DiscussionMaterials. Halbach, H. (1979): Khat - the roblem today. National Institute on Drug Abuse (NIDA) ResearchMonograph 27: 318- 19. ! Kalix, P. (1984): Recent advances inkhat research. Alcohol andAlcoholism 1914: 319-323. Kalix, P. (1984): The pharmacology ofkhat. GeneralPharmacology 1513: 179-189. Kalix, P. (1987): Khat: scientific knowledge and policy issues. Br. J. Addict. (Jan.) 8211: 47-53. Kennedy, J. (1987): The Flower of Paradise. The Istitutionalized Use of the Dru Qat inNorth Yemen.D.ReidelPublishingCompany, Dordrecht, Boston, Lancaster andTobo, 269 pp. Kennedy, J., J. Teague, W. Rokaw & E. Cooney (1983 : A medical evaluation of the use of qat inNorth Yemen. Social Scienceand Medicine 1 3112: 783-793. Lenaers, Anke and Peer Gatter. (2000). qat and its Impacts on NutritionalBalance and Family Life in Yemen: With a Special Reference to Low Income Segments o f Society, prepared for World Bank. Lemordant, D.(1983 :Can Khat be anything other than Harmful? In: ShahandehB.et al. (eds.): The Healthand ocioeconomic aspects ofKhat use. p. 139. B Mancioli, M.and A. Parrinello (1967): I1Qat (Catha edulis). La Clinica Terapeutica4312: 103- 172. Milich, L.and M. Al-Sabbry (1995). The "Rational Peasant" vs Sustainable Livelihoods: The Case of Qat inYemen. Development, Society for International Development. United Nations (1980): Bulletin on Narcotics: Special Issue Devoted to Catha edulis (Khat). Ward, C. & P. Gatter (2000): Republic of Yemen - A Qat Strategy. Report for the World Bank, Bulletin onNarcotics 3213. Ward,120PPA. al-Thawr, F. Qasem & A. No'man (1998): qat. Yemen Agricultural Strategy. C., Working PaperNo. 8, MinistryofAgriculture and Irrigation, Yemen. Weir,Publications. S. (1985): qat in Yemen, consumption and social change. London: British Museum World Health Organization (1980): Review ofthe pharmacology o f khat. WHO Advisory Group Bulletin onNarcotics 3213: 83-93. 28 Annex 1: The Qat Consumer Survey Sample In each governorate, between 400 and 600 individuals were interviewed (Annex Table 1). Overall, a total of 4,027 Yemenis above the age of 12 were interviewed, of which 1,807 were females (45%), and 2,220 males (55%). This was equivalent to 0.07% of the over-12 year of age population in the sample governorates2*. Of those interviewed, 1,909 persons or 47% were urban dwellers, while 2,118 persons or 53 YOlived in rural areas. Some 44% of the participating females were single, 55% were married and the remaining 2% were divorced, separated or widowed. Of the male participants, 33% were single, 59% were married and 8% were divorced, separated or widowed. Interms of educational achievement, some 12%of the male and 35% of the female respondentswere reportedto be illiterate. Every effort was made to ensure that the data was reliable. The survey was implemented by 72 locally-recruited surveyors, with an average o f 10 surveyors recruited per governorate. The surveyors were divided equally between men and women. They were formally trained prior to conducting interviews and were carefully supervised by an experienced field coordinator. Ineach district, random selection of householdswas undertakenafter approval to undertakethe survey was secured, with the assistance of local NGOs, from representativeso f the local security forces. While ensuringthat the sample was truly random, attention.waspaid to locating sufficient numbers of female and elderly respondents, to securing interviews in remote villages, and to reassuring respondents that aim of the survey was to gather information, and was not in any way relatedto recent radio broadcaststhat had discussedthe possibility of outlawing qal?. *'CSO StatisticalYearbook 2003, p. 29 and 30, and preliminary results of the 2004 Population Census, StatisticalYearbook 2004, p. 30. The survey coincided with broadcasts of al-Arabiyya and al-Jazira on qat in Yemen. These programs apparently voiced concerns about qat consumption and stated that a condition for Yemen to join the Gulf CooperationCouncil shouldbe the outlawing of qat. Respondentswere reassuredthat this survey was unrelated to the accessionto the GulfCooperationCouncil, or to any other effort aimedat banningqat use. 29 Annex 1, Table 1: Survey Areas and Number of Participantsby District and Gender Governorate District Gender Mate I Fcmak Total Sana'a Gov. I Saba'in 40 28 68 30 Annex 2: The RelationshipBetweenAge and Qat ChewingFrequencyfor Men and Women Annex 2, Figure 1: Age and ChewingFrequencyamongYemeni Males (Percenton left; age categories by color codes) 50 12-20 40 W 21-30 30 0 31-40 20 41-50 10 W 51-60 0 above 60 Source: World Bank Qat ConsumptionSurvey, April-May 2006, (Gatter 2006) Annex 2, Figure2: Age and ChewingFrequencyamongYemeni Females (Percenton left; age categories by color code) 90 80 70 60 50 0 31-40 40 30 041-50 20 W 51-60 10 0 above 60 Source: World Bank Qat ConsumptionSurvey, April-May 2006, (Gatter 2006) . 31 Annex 3: Household Expenditureon Qat Annex 3, Table 1: DistributionofHouseholdExpenditure (in Percent,2005) Source: HouseholdBudget Surveys 2005 Annex 3, Table 2: Household Expenditureson Qat By IncomeDeciles (YOof householdincome,2005) Deciles Urban Rural Total 1 6.7 8.8 8.6 2 8.2 9.3 9.1 3 7.8 10.0 9.6 4 8.9 10.1 9.8 5 9.5 10.7 10.4 6 9.6 11.2 10.8 7 10.1 11.1 10.8 8 10.2 10.7 10.6 9 10.9 11.6 11.3 10 11.2 10.5 11.0 Total 9.9 10.3 10.2 Source: HouseholdBudget Surveys 2005 32 33