Page 1 Report No. 39885 – EG Arab Republic of Egypt P OVERTY A SSESSMENT U PDATE (In Two Volumes) Volume II: Annexes September 16, 2007 Document of the World Bank Social and Economic Development Group Middle East and North Africa Region The World Bank Ministry of Economic Development Government of the Arab Republic of Egypt Page 2 Currency Equivalents (Exchange Rate as of September 10, 2007) Currency Unit = Egyptian Pound (LE) LE 1 = US$ 0.18 US$ 1 = LE 5.69 Fiscal Year July 1- June 30 V ICE P RESIDENT : D ANIELA G RESSANI C OUNTRY D IRECTOR : E MMANUEL M BI S ECTOR D IRECTOR : M USTAPHA N ABLI S ECTOR M ANAGER : M IRIA P IGATO T ASK T EAM L EADER : S HERINE A L -S HAWARBY Page 3 ANNEXES M ETHODOLOGY T ABLES F IGURES Page 4 T ABLE OF C ONTENTS A NNEX M ETHODOLOGY , D ATA AND S AMPLING .......................................................................................... 1 A NNEX 1.1: H OUSEHOLD I NCOME , E XPENDITURE , AND C ONSUMPTION , S URVEY - D ATA AND S AMPLING D ESIGN .................................................................................................... 1 A. HIECS Sample Design.. .................................................................................................................1 B. The HIECS Questionnaire:.............................................................................................................3 A NNEX 1.2: C OMMUNITY S URVEY ........................................................................................................ 4 A NNEX 1.3: E STIMATION OF H OUSEHOLD S PECIFIC P OVERTY L INE ..................................................... 5 A NNEX 1.4: D EVELOPING A P OVERTY M AP .......................................................................................... 7 A. The Consumption Model................................................................................................................8 B. Model Application ….. ....................................................................................................................9 A NNEX 2.1: A SSESMENT OF V ULNERABILITY TO P OVERTY …………………………………………..13 A NNEX 3.1: T HE E MPIRICAL F RAMEWORK OF E STIMATING THE W ELFARE I MPLICATIONS OF THE D EPRECIATION I NDUCED I NFLATION ............................................................................ 14 Step 1: Estimation of the Pass-Trough Effect………………………………………………………14 Step 2: Estimation of the Welfare Effect of the Changes in Prices Induced by Movements in the Exchange Rate…………………………………………………………….16 A NNEX 3.2: M ETHODOLOGY OF S IMULATING THE P OVERTY P ATH …………………………………. 17 A NNEX 4.1: E STIMATING H OUSEHOLD I NCOME P OVERTY .................................................................. 19 A. Identifying Household Characteristics Available in the HIECSs and the ELMSs ....................... 19 B. Estimating Per Capita Consumption Using the HIECSs Data…………………………………...19 C. Predicting Per Capita Consumption for the ELMSs Samples…………………………………....19 A NNEX T ABLES ......................................................................................................................................... 20 Table A.1.1: Daily Caloric Requirments by Age, Sex and Location.................................................. 20 Table A.1.2: Quantities and Calories Generated by the Reference Food Bundle............................... 20 Table A.1.3 Cost of 100 Calories by Region..................................................................................... 21 Table A.1.4: Sample Size of 1995/96, 1999/00 and 2004/05 Surveys ............................................... 21 Table A.1.5: Regression Results (Dependent Variable ln Household Expenditure), 1995/96, Metropolitan............................................................................................................ 22 Table A.1.6: Regression results (Dependent Variable ln Household Expenditure), 1995/96, Lower Urban ........................................................................................................... 23 Table A.1.7: Regression results (Dependent Variable ln Household Expenditure), 1995/96, Lower Rural ............................................................................................................ 24 Table A.1.8: Regression Results (Dependent Variable ln Household Expenditure), 1995/96, Upper Urban............................................................................................................ 25 Table A.1.9: Regression Results (Dependent Variable ln Household Expenditure), 1995/96, Upper Rural............................................................................................................. 26 Table A.1.10: Regression Results (Dependent Variable ln Household Expenditure), 1995/96, Border Urban........................................................................................................... 27 Table A.1.11: Regression Results (Dependent Variable ln Household Expenditure), 1995/96, Border Rural............................................................................................................. 27 Table A.1.12: Distribution of Poorest 50, 100 and 200 Sub-Districts by Governorate, 1996............ 28 Table A.1.13: Distribution of Poorest 50, 100 and 200 Sub-Districts by Governorate,2006............. 29 Table A.1.14: Distribution of Poorest 100, 500 and 1000 Villages by Governorate, 1996................ 30 Table A.1.15: Distribution of Poorest 100, 500 and 1000 Villages by Governorate, 2006................ 31 Page 5 Table A.2.1 (a): Poverty Measurements by Educational Attainment of Individuals, 2004-05........... 32 Table A.2.1 (b): Poverty Measurements by Educational Attainment of Individuals 2004-05............ 33 Table A.2.2 (a): Educational Status of Individuals by Region by Poverty Status, 2004-05............... 34 Table A.2.2 (b): Educational Status of Individuals by Region by Poverty Status 2004-05................ 35 Table A.2.3 (a): Poverty Measurements by Employment Status of Individuals, 2004-05 ................. 36 Table A.2.3 (b): Poverty Measurements by Employment Status of Individuals 2004-05.................. 37 Table A.2.4 (a): Employment Status of Individuals by Region by Poverty Status 2004-05 .............. 38 Table A.2.4 (b): Employment Status of Individuals by Region by Poverty Status, 2004-05............. 39 Table A.2.5 (a): Employment Status of Labor Force by Region by Poverty Status 2004-05............. 40 Table A.2.5 (b): Employment Status of Labor Force by Region by Poverty Status 2004-05............. 41 Table A.2.6 (a): Poverty Measurements by Sector of Employment of Individuals 2004-05.............. 42 Table A.2.6 (b): Poverty Risk by Sector of Employment of Individuals 2005................................... 43 Table A.2.7 (a): Sector of Employment of Labor Force by Region by Poverty Status 2004-05........ 44 Table A.2.7 (b): Sector of Employment of Labor Force by Region by Poverty Status 2004-05........ 45 Table A.2.8 (a): Poverty Measurements by Economic Activity of Individuals 2004-05.................... 46 Table A.2.8 (b): Poverty Measurements by Economic Activity of Individuals,2004-05 ................... 47 Table A.2.9 (a): Economic Activity of Labor Force by Region by Poverty Status 2004-05.............. 48 Table A.2.9 (b): Economic Activity of Labor Force by Region by Poverty Status 2004-05.............. 49 Table A.2.10 (a): Poverty Measurements by Employment Type of Individuals 2005....................... 50 Table A.2.10 (b): Poverty Risk by Employment Type of individuals and by Region, 2005.............. 51 Table A.2.11 (a): Type of Employment of Individuals in Labor Force by Region by Poverty Status 2004-05.................................................................................. 52 Table A.2.11 (b): Type of Employment of Individuals in Labor Force by Region by Poverty Status 2004-05................................................................................. 53 Table A.2.12 (a): Poverty Measurements by Household Size 2004-05............................................. 54 Table A.2.12 (b): Poverty Measurements by Household Size 2004-05............................................. 55 Table A.2.13 (a): Distribution of Individuals by Household Size, by Region and by Poverty Status 2004-05......................................................................................... 56 Table A.2.13 (b): Distribution of Individuals by Household Size, by Region and by Poverty Status 2004-05........................................................................................ 57 Table A.2.13 (c): Poverty Risk of Households By Number of Children, by Region and Poverty Status, 2005................................................................................................. 58 Table A.2.14: Average Household Size by Poverty Status for 2004-05 and 1999-00........................ 59 Table A.2.15 (a): Demographic Characteristics by Poverty Status and Region 2004-05................... 59 Table A.2.15 (b): Demographic Characteristics by Poverty Status and Region 2004-05.................. 60 Table A.2.16: Poverty Measurements by Household Structure and Gender of Household Head, 2004-05.......................................................................................... 61 Table A.2.17: Distribution of Individuals by Household Structure, by Gender of Household Head and by Poverty Status, 2004-05 ........................................ 62 Table A.2.18 (a): Poverty Measurements by Gender of Household Head, 2005................................ 63 Table A.2.18 (b): Poverty Risk by Gender of Household Head, 2005............................................... 64 Table A.2.19 (a): Distribution of Individuals by Gender of Household Head, by Region and Poverty Status, 2005............................................................................... 65 Table A.2.19 (b): Distribution of Individuals by Gender of Household Head, by Region and Poverty Status, 2005.............................................................................. 66 Table A.2.20: Illiteracy Rate among Children of Age 12-15 Years Old by Poverty Status and Region 2004-05 .............................................................................. 67 Table A.2.21: Percentage of Working Children Aged 6-15 Years by Poverty Status and Gender, 2004-05........................................................................................... 68 Table A.2.22: Net Enrolment Rate in Basic Education by Poverty Status and Gender 2004-05........ 69 Table A.2.23: Shares of Different Income Sources by Poverty Status and Gender of Household Head 2004-05............................................................................. 70 Page 6 Table A.2.24: Percentage Shares of Different Types of Transfers , Out of Total Income, by Poverty Status and Gender of Household Head 2004-05 ........................... 71 Table A.2.25: Percentage of Households with Public Amenities Characteristics by Poverty Status 2004-05.................................................................... 72 Table A.2.26: Percentage of Households by Ownership of Durable Goods and by Poverty Status 2004-05...................................................................................... 73 Table A.2.27: Share of Various Expenditure Items to Total Expenditure by Poverty Status 2005.... 74 Table A.2.28: Fertility Rate and Under Five Mortality Rate by Poverty Status, 2004-05.................. 75 Table A.2.29: Unemployment Rate of Youth (15-24 years) by Educational Status and Poverty, 2005............................................................................................... 75 Table A.2.30: Net Enrolment Rate by School Type and Poverty Status for Different Levels of Education, 2004-05......................................................................... 76 Table A.2.31: Regression of Log Welfare Measure (Consumption/Poverty Line) on Characteristics of Household and Household Head for 2004-05 and 1999-00.......... 77 Table A.2.32: Impact of Changes in Household Characteristics and Characteristics of the Household Head on Poverty........................................................ 78 Table A.3.1: Exchange Rates and Consumer Prices, 2000-2005........................................................ 79 Table A.3.2: Disaggregated Price Change.......................................................................................... 79 Table A.4.1: Estimated Per-Capita Region-Specific Poverty Lines (L.E. Per Year) for 1999/2000 and 2004/2005................................................................................. 80 Table A.4.2: Employment Structure and Growth Rate by Type of Employment, Sex and Urban/Rural Location, 1998-2006................................................ 81 Table A.4.3: Employment Structure and Growth Rate by Economic Activity, Sex and Urban/Rural Location 1998-2006....................................................................... 82 Table A.4.4: Cross-Sectional and Longitudinal Method of Calculating the Growth in Agriculture Wage and Agriculture Non-Wage Work by Sex and Urban/Rural Location 1998-2006.................................................................. 83 Table A.4.5: Distribution of Real Monthly Earnings for Wage and Salary Workers by Background Characteristics, 1988-2006...................................................... 84 Table A.4.6: Distribution of Real Monthly Wage for Wage and Salary Workers by Institutional Sector and Economic Activity, 1998-2006............................................. 85 Table A.4.7: Share of Low Monthly Wage Earners, Wage and Salaried Workers 1998-2006 .......... 86 Table A.4.8: Transition Across Low/High Earnings by Sex, 1998, 2006 from Wage Employment in 1998 to Wage Employment in 2006 ................................... 87 Table A.4.9: Transition across Low/High Earnings by Institutional Sector, 1998, 2006................... 87 A NNEX F IGURES ....................................................................................................................................... 88 Figure A.1.1: Predicted Poverty Rates at Village Level and Their Confidence Intervals, in Rural Areas, 1996......................................................................................... 88 Figure A.1.2: Predicted Poverty Rates at the Sub-District Level and Their Confidence Intervals, in Urban Areas, 1996..................................................................... 88 Figure A.3.1: Distribution of Estimated Long-Run Exchange Rate Pass-Through to Consumer Prices........................................................................................................... 89 Figure A.3.2: Direct Effects of Price Changes on Welfare (Compensating Variation Calculated as Percent Change in Total Expenditure Required to Purchase Initial Consumption Basket)......................................................................... 89 Page 7 Figure A.4.1: Distribution of Real Monthly Earnings in Relation to a Low Earnings Threshold by Sex, 1998-2006 (Using CPI)....................................................... 90 Figure A.4.2:Distribution of Real Monthly Earnings in Relation to a Low Earnings Threshold by Institutional Sector of Employment, 1998-2006 (Using the CPI)............... 91 Figure A.4.3: Distribution of Real Monthly Earnings in Relation to a Low Earnings Threshold, 1998-2006 (Using the FPI)............................................................... 92 Figure A.4.4: Distribution of Real Monthly Earnings in Relation to a Low Earnings Threshold by Institutional Sector of Employment, 1998-2006 .......................... 93 Page 8 A NNEX M ETHODOLOGY , D ATA AND S AMPLING A NNEX 1.1: H OUSEHOLD I NCOME , E XPENDITURE , AND C ONSUMPTION S URVEY - D ATA AND S AMPLING D ESIGN Egypt conducted household budget surveys since 1957/58. It was intended to perform these surveys every five years. But because of .unavailability of funds, these surveys were stopped for some time. Dates for these surveys are 1957/58, 1964/65, 1974/75, 19981/82, 1990/91, 1995/96, 1999/2000 and 2004/2005. Household Income, expenditure and consumption surveys (HIECS) present the single most important source of information for poverty analysis. They record information on household income and consumption expenditures on more than 600 items of goods and services, and are therefore a good source of information on the distribution of welfare within the society. These surveys are particularly important because of their comparability, in terms of survey design and administration, and hence the opportunity they offer in making comparisons over time. However, the three surveys are slightly different in terms of sample selection and topics covered by the questionnaires. But differences do not affect comparability of them. A. HIECS S AMPLE D ESIGN The samples of the three surveys are stratified multistage random samples. The sample designs of all surveys were nationally representative and the size for both surveys is large enough to allow for inferences at the regional and governorate levels, with the exception of Border governorates where the sample size is small. Levels of bias and imprecision for both surveys are within statistically acceptable margins. Using the variance and mean expenditure of previous survey, it was estimated that the sampling errors in the 1999/2000 survey were 0.7 percent in urban areas and 0.9 percent in rural areas, with 95 percent confidence level. The sample design is stratified, multistage design can be explained as follows: The master sample is stratified such that urban and rural areas are self-independent strata. Each strata (urban or rural) is divided into internal layers (being the governorates), with probability proportion to size from an updated population Census of the closest year. PSU’s (areas) were systematically selected, using sampling interval and a random start. Using maps, these areas were further subdivided into a number of chunks of about certain number of households each and one chunck is chosen randomly from each area. Household lists for the selected Chuncks were prepared. Finally, households were selected randomly from each chunk. Sampling design of 2004/05 Survey The 2004/05 HIECS sample is multi stages self weighted area sample of 1223 PSU of about 700 household each. Total PSUs were distributed among urban and rural areas using proportion to size criteria, and then Urban and rural PSUs were distributed proportionally between governorates. Thus each governorate is represented in the master sample; however the number of PSUs in Border governorates may be very small. 1 Page 9 2 Selection of Primary sampling units The first sampling stage is selecting a sample from villages from rural areas frame and Shiakh (or part of it) and capital of Markaz(district) from urban areas frame. Master sample of 1223 PSUs was distributed between urban and rural stratum such that the share of each stratum in PSUs equals its population share and assuming that there are 600 households in each PSU according to 1996 Census. However, some small villages were pooled together to ensure the required size of PSU (600 households). In urban areas, sub districts were arranged geographically using zigzag method to ensure balanced spread of the sample within each governorate. While, in rural areas illiteracy rate was used to arrange villages, where village in the first Markaz are arranged in descending order, then villages in the second Markaz were arranged ascending, and so on. Systematic sampling criteria was used in this stage. The selected villages or sub districts were divided into small areas of similar number of households of 600, according to 1996 population Census, and then one area is chosen at random from each PSU. A list of all households within the selected area unit was prepared, where quick count showed that every selected area include about 700 households. A sample of 40 households was selected randomly from each area sampling unit. Although the 1999/2000 and 1995/96 sampling designs were similar to that of 2004/05 sample, there are some differences; first the sample is self weighted within Urban and rural stratum but not at the national level; second the number of PSUs in 600 in 1999/00 and 500 in 1995/96, where 80 and 30 households were randomly chosen from each PSU in 1999/00 and 1995/96 respectively., see table A1.1.1 for sample size and distribution; and 1999/2000 and 2004/2005 HIECS were based on the1996 Census sample frames while 1995/96 sample was based on a 1993 update of 1986 Census data. One interesting characteristic of the sample selection method is that all governorates in urban and rural areas are represented in each quarter (three successive months), thus sample surveyed during each quarter is also nationally represented and therefore no seasonal bias can be detected in any areas. Table A1.1.1 Sample Size of 1995/96, 1999/00 and 2004/05 Surveys Data of the 2004/05 survey was collected from July 2004 to June 2005, while data for 1999/00 and 1995/96 was collected from October of 1999 and 1995 to September 2000 and 1996, respectively. 1995/96 1999/2000 2004/2005 Number of Households Number of Individuals Number of Households Number of Individuals Number of Households Number of Individuals 6622 28911 28754 125287 21743 88843 8183 45028 19195 100830 25352 118588 14805 73939 47949 226117 47095 207431 Page 10 3 B. T HE HIECS QUESTIONNAIRE : The survey was administered over 12 months, with 10 visits to each household over a period of one month. This is the largest survey ever conducted in Egypt. The last three surveys of 1995/96, 1999/2000 and 2004/05 are highly comparable in terms of data collection procedures. The measure of total consumption used in this report is quite extensive and draws upon responses of several sections of the survey. Two survey forms were used in HIECS, a diary and a main questionnaire. Each household was visited ten times over the course of one month. The enumerator gave the household a diary in the first visit and asked the respondent to report each of the food expenditure items that the household makes every day, for a period of one month. The sum of the daily expenditure was then recorded in the main questionnaire at the end of the interview cycle. Expenditure of non food items were collected for the previous three month or the previous year depending on the type of commodity. The annualized sum of monthly or quarterly household expenditures was then used to construct the consumption basket for total annual household expenditures. Interviewers took down household demographic information at the first interview and household income at the last two interviews. In brief, consumption is measured as the total sum of food consumption (home produced and markedly purchased), total non food expenses, an actual or imputed rental value of housing. The questionnaire consists of seven sections on a series of topics which integrate monetary to non monetary measures of household welfare and a variety of household behavioral characteristics. The first section is concerned about the basic information of all household members such as age, sex, relation to head of household, education and employment status. In the second section information on housing and basic amenities are collected. Possession of durable goods is reported in section three. Food consumption includes food which the household has purchased, grown and received from other sources for 279 items, where these data are reported in section four. Non food consumption is the sum of expenditure of 298 non food items, including expenditure on fuel, clothing, schooling, health, and several miscellaneous items. Information on consumption on non food goods and services is registered in section five. Section six is concerned with Transfer and credit expenditure, while income by detailed income sources is obtained from special income questionnaire. Although the three surveys follow the similar format almost exactly and total consumption definitions and recall periods are similar in all survey years, additional important information was collected in 2004/05 survey. Namely, first: in kind received goods were reported separately, second: information on school enrolment and household education expenditure on public or private education were reported, third: evaluation of the existing assets and changes in them were reported to allow for evaluating savings and dis-savings, and forth; the household questionnaire was supplemented by a community questionnaire as will be discussed below. In terms of quality, the survey data can be judged “better than average”. The samples are nationally representative. They were randomly and systematically chosen, and a stratified multiple stage sampling was used. The sample size for the survey is large enough to allow for inferences at the regional and governorate levels, with the exception of Border governorates where the sample size is small. Levels of bias and imprecision for the survey are within statistically acceptable margins. Page 11 4 A NNEX 1.2: C OMMUNITY S URVEY Integrated with HIECS, community data were also collected for all communities of PSUs in CAPMAS master sample of HIECS. Community data provided by the Community Survey include data on water and sewerage systems, health posts and schools and quality of agricultural land and main crops grown. The Community Survey was administered in all 1223 PSUs of the CAPMAS master sample. However, satellite villages were considered as separate communities and thus the total number of communities in the rural sample was 1095 communities (mother or satellite villages) rather than 675 PSUs. Besides, there is no clear distinction between sub-districts (shiakha) in urban areas, so it was decided to collect information at the district (kism) level in urban areas. Thus, the total number of communities is 1390 communities from the master sample of CAPMAS. The community questionnaire covers the following areas: 1- Availability, accessibility, and quality of facilities in the community such as schools, health units, police stations, etc.; 2- Availability, accessibility, and quality of infrastructure in the community such as potable water, electricity, sewerage system, etc.; 3- Information on SFD interventions and other community interventions; 4- Perceptions on community participation in the project. This section sought to characterize the community’s participation in the project cycle. Were they \03 consulted? How? Were they able to make decisions? What type of decisions? Would they use \03 the facility? 5- Community perceptions on the impact of the project. This component sought to establish the com munity’s perception of the benefits and \03 disadvantages of the project. The participants were asked to evaluate the priority (i.e. relative \03 importance) and usefulness of the project, the quality of the installation, the benefits at \03 household and community levels, and those received by neighboring communities. Page 12 5 A NNEX 1.3 E STIMATION OF THE H OUSEHOLD S PECIFIC P OVERTY L INE The report follows the cost of basic needs methodology to construct household region-specific poverty lines. This methodology, which was adopted also in the World Bank 2002 report, takes into account : (i) ‘economies of scale’ within households – the fact that non-food items can be shared among household members; (ii) differences in non food consumption patterns and prices across regions in Egypt; and (iii) differences in ‘basic needs’ requirements of different household members – young versus old, male versus female . For consistent poverty comparisons, this report adopts the same methodology in estimating food and non food basic needs. This method is outlined below. It was preferred to use different food baskets that reflect the consumption preferences of the second quintile of the year under consideration rather than using one food basket for both years and evaluated at the corresponding prices. In fact the period of 2000-2005 exhibited large increase in food prices following Egyptian pound devaluation on 2003, and the Government of Egypt responded to this change by subsidizing several pulses and grains. Price changes and subsidizing food items that are largely consumed by the poor, may have caused changes in consumption patterns, thus we preferred to use different food baskets that reflect consumption behaviour of the second quintile. H OUSEHOLD - SPECIFIC POVERTY LINES Differences in poverty lines reflect variations in the food and non-food prices across the seven regions. They also incorporate household differences in the size, gender and age composition, and their food and non-food consumption preferences. 1. Caloric Requirements The FAO has been concerned with the issue of determining the nutritional norms of individuals in different age and sex groups. These norms vary from country to country (and even amongst different groups within a country). Nutritional needs of individuals are the starting point to construct food poverty lines. It must be emphasized that these needs of individuals depend on several factors such as age, sex, location conditions and activity levels. We first estimate minimum caloric requirements for different types of individuals. Using tables from WHO, caloric needs are separately specified for urban and rural individuals, by sex and 13 age categories. For individuals over 18 years of age, WHO's recommended daily allowances are differentiated by weight and activity levels. The estimates used in this paper assume the average weight of men over 18 years of age is 70 kg and 60 kg for women. Urban individuals are assumed to need 1.8 times the average basal metabolic rate and rural individuals are assumed to need 2.0 times average BMR. Thus, each household has its own caloric requirements depending on its location, age, gender decomposition, table (A.1). 2. Food Poverty Line Once the minimum caloric needs have been estimated, the next step is to determine the cost of obtaining the minimum level of calories. Cost is determined by how the calories are obtained on average by the first two quintiles, rather than by pricing out the cheapest way of obtaining the calories or following a recommended diet. For the first two quintiles of households ranked by nominal per capita consumption, average quantities of all food items are constructed. Total calories generated by this bundle are calculated using calories contents in every food item. Table A.2 demonstrates quantities and calories generated by the reference food basket. These quantities Page 13 6 represent the bundle used to estimate the food poverty lines, which reflect consumption preferences of the poor. The bundle was priced using unit prices prevailing in each region. Dividing cost of the chosen bundle by calories generated by it, the costs per calorie in each region were obtained Household specific food poverty line is derived by multiplying calorie requirements for all household members by relevant cost of calories. Food poverty line takes into account household gender and age composition as well as its residential region. Food poverty line is used define extreme poverty, where households whose total actual consumption are below their food poverty lines, are considered ultra poor. This stage can be explained mathematically as follows: let Z denote the actual food consumption vector of the reference group of households initially considered poor; first two quintiles. The corresponding caloric values are represented by the vector k, and the food energy intake of the reference group is then k z = k.Z'. Let cost of this bundle for region r is P r , and caloric requirements of household h is C h , thus the cost of one calorie in region r is given by ( k z /Pr) . Food poverty line for household h is then given by ( k z /Pr)* C h, thus the relative quantities in the diet of the poor are preserved in setting the poverty line. Table A.3 shows regional cost of 100 calories generated by the reference bundle. 3. Non food Poverty Line While the cost of the minimum food bundle is derived from estimated physiological needs, there is no equivalent methodology for determining the minimum non- food bundle. Following Engel’s law, food shares are regressed against logarithm of total household expenditure relative to food poverty line and its square, logarithm of household size and its square, share of small and older children, share of adult males and females, and share of elderly. That is i h f z i x f z i x i s GJ E D \0e\0e \0e 2 )) / (log( ) / log( , (1) Where s i denotes food share of household i, x i is its actual consumption, z f if the food poverty line and h i is vector of household demographic characteristics. The non-food allowance for each household can be estimated in two ways: (i) regressing the food share against total expenditures and identifying the non-food share in the expenditure distribution of households in which expenditure on food is equivalent to the food poverty line; or (ii) by identifying the share of non-food expenditure for households in which total expenditure is equivalent to the food poverty line. The former approach yields an “upper” bound of the poverty line, while the latter yields a “lower” bound, since it defines the total poverty line in terms of those households which had to displace food consumption to allow for non-food expenditures, considered to be a minimum indispensable level of non-food requirements. Thus lower poverty line =(2- S i )*Z f (2) . Upper poverty line is obtained by solving equation (1) iteratively. By this approach household regional specific poverty lines are estimated (households with the same gender and age composition in each region have the same poverty lines). Obviously this approach takes into account location, age and gender composition as well as economies of scale; as food shares and hence non food estimates vary according to household size, age and gender composition. Hence differences in food shares result from the addition of members of specific age and gender. The sharing behaviors among household members are also reflected. Page 14 7 A NNEX 1.4 D EVELOPING A P OVERTY M AP Poverty maps, as developed by Elbers et al. (2002), are based on a statistical procedure that combines both household survey data with population census data. On the one hand, household surveys provide statistically reliable spatial estimates of consumption (as an indicator of welfare) at the regional level, separately for urban and rural areas. On the other hand, the extensive coverage of the census, which does not contain any information on consumption, provides more disaggregated data. The Elbers et al. (2002) statistical procedure also allows for heteroskedasticity in the household component. In a nutshell, survey data are first used to estimate a prediction model for consumption and then the parameters are applied to census data to derive an imputed value for consumption, employing a set of explanatory variables which are common to the survey and the census. This allows defining a set of welfare indicators based on consumption such as headcount poverty. Finally, the welfare indicators are constructed for geographically defined subgroups of the population using these predictions. Although the approach is conceptually simple, properly accounting for spatial autocorrelation in the first stage model and estimating standard errors for the welfare estimates requires additional elaboration. Although the approach is conceptually simple, properly accounting for spatial autocorrelation in the first stage model and estimating standard errors for the welfare estimates requires additional elaboration. The Method in Details: The method in this study can be thought of as being divided into three stages that occur in sequence. The three-stage procedure is implemented using HIECS 1995/96 data and Census 1996 data. The first stage in the poverty mapping exercise involves a rather painstaking comparison of common explanatory variables across the household survey and the population census. A concurrent exercise that was carried out in parallel to the exercise described above is the compilation of a database at a level of aggregation higher than the household, which can be inserted into the household level census and the household survey databases. The two tasks described above yield a good and reasonably large set of common household-level variables, supplemented by a series of additional variables at a slightly higher level of disaggregation. A set of common variables to both the survey and the census is selected. Using the household survey and the variables selected in the first stage, the second stage analysis involves the econometric estimation of models predicting household consumption on the set of household-level and community variables. Each region was treated separately thus seven models are built for the seven regions. Tables A1.5 to A.1.11 provide the semi-log models of household expenditure in the seven regions of Egypt based on HIECS 2004/2005 Survey. Successful completion of the second-stage analysis permits one to take the parameter estimates and attendant statistical outputs to the third stage. The estimated parameters are transferred to the data from the population census to simulate the consumption level of each and every household enumerated in the population census. The simulated household consumption is then used as the basis for calculating poverty and other welfare indicators at a variety of levels of spatial disaggregation. Statistical precision of the welfare estimates is also gauged in this stage. Finally, the welfare indicators are constructed for geographically defined subgroups of the population using these predictions. Once the poverty map exercise has been completed for all regions in the country, the resulting databases which provide estimates of poverty and inequality (and their standard errors) at a variety of levels of Page 15 8 geographic disaggregation can be projected onto geographic maps using GIS (Geographic Information Systems) mapping techniques. When inspecting these maps it should be kept in mind that they have been created using the expected headcount. The true headcount for a location will differ from the expected headcount because of sampling and modeling error. One of the key advantages of poverty maps technique is that estimates of welfare are obtained, but also standard errors associated with those estimates are derived. A general impression of overall precision levels can be gauged from Figures 4.4 and 4.5. To show what precision can be achieved at the sub-district level/ village, Figures 4.4. and 4.5. show the village/sub- district level predicted poverty headcount, using along with brackets giving confidence interval; two standard error below and two standard error above the point estimate. A. T HE C ONSUMPTION M ODEL To estimate household consumption levels, a standard reduced-form framework in which log household consumption is regressed on household characteristics, including human and physical capital, as well as on community-level characteristics. Community characteristics are specified at the village level. The HIECS contains a limited but important set of variables that can be used to explain households' consumption levels. Variables include demographic variables, variables that characterize the household's human capital such as literacy rates, employment variables such as agricultural employment, housing characteristics and the availability of basic amenities. The final specification included only those variable that were significant at least at 95 per cent. The resulting residuals are then checked to see if there are some outliers in the observation. The location residuals were then regressed on a set of census means at village level. A selection criterion of significance at 95% was applied. Following the inclusion of these additional variables, the GLS (Generalized Least Squares) regression was re-estimated in order to reduce the size of the location effect. Following Elbers et al. (2001, 2002), the empirical model of household consumption is defined as ch ch ch ch u y E y += ] [ln ln x (1), where Ln ch y is the logarithm of consumption of household h in village c, ch x is a vector of observed characteristics of this household (including village level variables), and ch u is the error term. Note that ch u is uncorrelated with ch x . This model is simplified by using a linear approximation to the conditional expectation ] x [ln ch ch y E and decomposing ch u into uncorrelated terms ch c ch u e h + = (2), where c K represents a village level error term common to all households within the village, and ch H is a household specific error terms. It is further assumed that the c K are uncorrelated across villages and ch H are uncorrelated across households. With these assumptions, equation (1) reduces to ch ch ch u y + = x ' ln (3). Estimation of the parameters underlying this equation, in particular the vector of parameters and the distributional characteristics of the error terms, can be done by using standard tools from econometric analysis (see Elbers et al. , 2002). The consumption model specification in equation (3) allows for an intra-village correlation in the error terms. Household income or consumption is certainly affected by the location where the Page 16 9 household lives. Even though ch x has some variables representing village level characteristics, it is quite plausible that some of the location effects will remain unexplained. The consequence of failing to take into account this within-village correlation of the error terms can result in biased welfare estimates and will generally lead to underestimation of standard errors of welfare estimates. As mentioned above, the estimate of c K for each cluster in the census dataset is not applicable, therefore we must estimate the deviation of c K . Taking the arithmetic expectation of (3) over village c . . cc c u HK \0e (4), hence . Assuming c K and ch H are normally distributed and independent each other, Elbers et al gave an estimate of variance of the distribution of the location effect c K : (5). When the location effect c K does not exist, equation (3) is reduced to ch ch u H . B. M ODEL A PPLICATION This section outlines the stages and procedures implemented in applying the model to obtain poverty maps for governorate, district, sub district and village/city levels. Five models were estimated, separately, for the five regions. Stage 1: Matching Variables in the Survey and the Census In order to obtain rigorous estimates of consumption levels of the households in the census, the explanatory variables selected in the consumption determination model have to exist and are measured in the same way in both the household survey and in the census. If the sample of the household survey was randomly selected and nationally representative, the distribution of each explanatory variable in the household survey can be expected to be the same as its distribution in the census. Stage 2: Selecting Explanatory Variables for the Consumption Model The procedure in selecting the explanatory variables of equation (3) starts by running a regression of log consumption on the matched variables identified in Stage 1, plus some variables that can be created from those variables such as the square of household size or the square of age of household head. In order to obtain a robust specification, variables are only selected for inclusion in equation (3) if they contribute significantly to the explanation of (log) household consumption. Hence variables with low t-values are dropped. After the appropriate set of variables has been selected in this way, the regression is run again and the residuals of this regression are saved. These residuals need to be scrutinized to check if there are some outliers in the observation. If indeed there are some residual values which are far out of the Page 17 10 range of most residual values, then these observations must be checked for coding or other errors. Ultimately it may be necessary to delete them from the data. Fortunately, this is extremely rare. The next step is to select village-level independent variables to complete the consumption model specification. The village level variables are obtained from either the census data aggregated at the village level (for example the total number of individuals in the population or means of age of household heads in each village). These variables are then grouped into several sets such as demographic variables, village infrastructure variables, and village economic variables. The residuals of the last regression are then aggregated at the village level to calculate the mean of these residuals for each village. The variable selection is then done by running separate regressions of the village-level mean of residuals on each set of the village-level variables. The variables with significant t-values are selected as the candidates for inclusion in the consumption model. Stage 3: Estimating the Consumption Model The result of stage 2 is a complete specification of the consumption model, incorporating both household-level and village-level independent variables of the model. The next step is to test whether there is heteroskedascity in the data. This will determine the method to be employed to estimate the model. The first step to do this is to estimate the model of equation (3) using Ordinary Least Squares (OLS) and save the residuals as a variable ch u \0c . Based on equation (2) the residuals ch u \0c are then decomposed into uncorrelated components as ch e c ch u \0e K ˆ ˆ . To investigate the presence of heteroskedasticity in the data, a set of potential variables that best explain the variations in ch e 2 are used to estimate the following logistic model (6) where A set to equal 1.05*max } 2 { ch e as in Elbers et al. , (2002). This specification puts bounds on the predicted variance of } 2 { ch H . In the case where homoskedasticity is rejected, a household specific variance estimator for ch H is calculated as (7) where } ˆ { D T ch Z EXP B . The result from above indicates a violation of assumptions for using the OLS in model (3), so a GLS regression is needed. In GLS the variance-covariance matrix is a diagonal block matrix with structure: ๚ ๚ ๚ ๚ ๚ ๛ ๙ ๊ ๊ ๊ ๊ ๊ ๋ ้ + + + + e h e e e e e h e e e e e h e e e e e h s s s s s s s s s s s s s s s s s s s s c c c c (8). Overall, the procedure for estimating the consumption model of the poverty mapping computation Page 18 11 can be listed as: 1. Estimate model (3). 2. Calculate the location effect c K (2). 3. Calculate the variance estimator ) 2 var( K V (4). 4. Prepare the residual term ch H for estimating model (6). 5. Estimate GLS model with (8). 6. Use a singular value decomposition to break down the variance-covariance matrix from previous step. This will be used for generating a vector of a normally distributed random variable such that the joint variance-covariance matrix will be in the form of (8). 7. Read in census data, eliminate records containing missing values, generate all census variables needed for models (2) and (6). 8. Save all datasets needed for the simulation. Stage 4: Simulations on Census Data The purpose of this procedure is to apply the parameters estimated in the previous procedure to the census data. However, since the values of these parameters are obtained through estimations, they are not the precise values of these parameters and subject to sampling error. This needs to be taken into account in applying the parameters to the census data by taking into account the sampling error of the coefficient estimates. To start, recall that the purpose is to calculate the simulated version of equation (3): s ch s c s ch ch s y HK \0e \0e ' x ln , (9) where the superscript s refers to simulated version of each parameter or variable and now ch x refers to characteristics of the households in the population census data. Simulation of E The simulated value of E is obtained through a random draw, assuming ) ˆ , ˆ ( ~ E E E 6 N . Note that the draw has to take into account the covariance across E ’s . The randomly drawn parameter is defined as s E . The next step is then to apply this simulated parameter to each household in the census data to calculate the value of s ch x E . Simulation of c K The process of obtaining the simulated value of c K requires two steps of simulations. This i s because the variance of K itself is estimated with error. Hence, the first step is to obtain the simulated Page 19 12 variance of s 2 , K V K . Elbers et al. (2002) propose to draw s 2 K V from a gamma distribution )] 2 ( ˆ , 2 ˆ [ ~ 2 P V P V K V ar V G . Accordingly, a random draw of the variance for the whole sample is exercised and its mean is defined as s 2 K V . Then the second step is to randomly draw s c K for each village in the census data, assuming ) 2 , 0 ( ~ s c N c V K . Simulation of ch H The process of obtaining the simulated value of ch H requires the use of the results of estimation of equation (6). Assuming )] ˆ ( , ˆ [ ~ DD D Var N , a random draw of D is made and defined as s D . Like in the case of E , the draw has to take into account the covariance across D ’s . The simulated parameter is then used to simulate the household specific variance estimator for ch H as defined in equation (7) for each household in the census data. Finally, the simulated value of household specific idiosyncratic shock, ch s H for every household in the census data is obtained by taking a random draw, assuming ) 2 , 0 ( ~ s ch N ch V H . Collecting Now all the three components of equation (9) have been simulated, the value of ch s y ln for all households in the census data can be calculated by summing up the values of s ch s c s ch H K , , ' x that have been obtained. The whole set of simulations is then repeated a number (100) of times, so that in the end a database of 100 simulated values of (log) household expenditure of all the households in the census data is created. Stage 5: Calculation of Poverty and Inequality Indicators The final output of stage 4, a database of 100 simulated values of household expenditure of all households in the census data, is used as the basis for calculating various poverty and inequality measures at the provincial, district, sub-district, and village/city levels. The point estimate of each measure is the mean of the calculated measure over the 100 simulation values. Meanwhile, the standard error of this estimate is equal to the standard deviation of the calculated measure over the 100 simulation values. Page 20 13 ANNEX 2.1 A SSESSMENT OF V ULNERABILITY TO P OVERTY To assess the vulnerability of households in Egypt we rely on a two-step approach. Let welfare measure (total household consumption deflated by poverty line). W i be a function of household characteristics X i and assume that W i is log-normally distributed. In the log form: ln(W i )=X i b + e i , (1) where e i is a normally distributed error term. Then the probability of household i to be poor, or, in our definition, the vulnerability of household i is V i =prob(ln(W i )<0))= F ((- X i b )/ s ), (2) where F is a standard normal distribution function. Thus, in the first stage we model the determinants of household consumption in the form of equation (1). In the second stage, we simulate the effect of the covariates from the consumption regression on the probability that a household will be poor. The poverty profile presented in the previous section provides guidelines for the selection of the potential variables to be included in this regression. It was assumed that vulnerability can be affected by four sets of variables : education, employment, demography and housing characteristics, such that policy implications of educational investments, employment patterns, and investment in family planning can be evaluated. The set of explanatory variables includes household size, household demographic variables, shares of individuals with university degrees and illiterate household members, share of unemployed, characteristics of the household h ead that include gender, age, age squared, and a set of dummies for the head’s educational level as well as working status and sector of employment. We run separate regressions for four urban and three rural regions of Egypt. Similar to Datt and Jolliffe (1998) we use a fixed effect regression specification on region level to correct the bias in the estimated coefficients due to potential endogeneity or omitted variable bias. Local characteristics, such as the degree of infrastructure development, geographical location, fertility of land, etc., while not registered in our data, might affect the level of consumption of the households living in particular locality. Omitting these variables in our specification could lead to inconsistency of parameter estimates. The fixed effect specification should control for this type of omitted variable bias. Estimation of Household Welfare Two models were estimated for 2004/2005. The first includes households' socio economic characteristics as well as housing conditions, while in the second model, the characteristics of communities where households live, were also used as explanatory variables. The first model was used to compare results of 2004/05 survey with 1999/2000 and 2004/05 surveys. Table A.2.31 demonstrates the results of regression for years 1999/2000 and 2004/05. Most of the variables were highly significant (except the gender and out of human force variables) and adjusted R2 exceed 50 percent. Table A.2.32 shows the mean and standard deviation of all the variables included in the model. Page 21 14 ANNEX 3.1 T HE EMPIRICAL FRAMEWORK OF ESTIMATING THE WELFARE IMPLICATIONS OF THE DEPRECIATION INDUCED INFLATION A two-step methodology was adopted. The first step consists of isolating the component of observed price changes during this period that are due to the depreciation for 160 different price indices 1 . Disaggregation of the exchange rate pass-through to this level is important, as there is considerable heterogeneity across commodities in the response of domestic consumer prices to the exchange rate. The second step was to then bring the estimated price changes due the depreciation for each of these 160 different price indices to the household survey for Egypt, to investigate their welfare effects. S TEP 1: E STIMATION OF THE P ASS -T ROUGH E FFECT Monthly disaggregated data on the consumer price index are used to calculate “pass through estimates”. Because of difficulties in mapping the expenditure items in the CPI to the household survey, a somewhat more aggregated set of 20 of these expenditure items that correspond to expenditure categories in the household survey are used. The consumer price of item i in region r in month t is modeled as follows: (1) ( ) ( ) ir ir 1 T irt N irt irt P P P a - a ื = where P N denotes the price of the non-traded component and P T denotes the price of the traded component of that item. To simplify notation, the non-traded component captures both purely non- traded goods within this item, as well as non-traded distribution costs associated with the traded goods. Accordingly P T is considered as the price of imported goods "on the dock" in Egypt. Concretely, one of the disaggregated items is fruit. P T would therefore be the price of imported fruit "on the dock", while P N is a price index of non-traded fruit as well as the distribution costs associated with both kinds of fruit. Following the large empirical literature on exchange rate pass-through, the logarithm of this import price is modeled as a linear function of the log exchange rate and a measure of foreign marginal costs of production: 2 (2) irt t ir 2 t ir 1 ir 0 T irt u C ln ) L ( E ln ) L ( P ln +ื b + ื b + b = where E is the exchange rate, C is a proxy for foreign marginal costs, and u is an error term that I assume is independent of the exchange rate. No direct measure of foreign marginal costs of production disaggregated by product is available. An aggregate foreign cost variable, which is a trade-weighted average of the monthly producer price index in Egypt's five largest trading partners for which this data exist, is introduced. 3 The extent of foreign cost pressures on export prices is made to vary by product and region. b 1 (L) and b 2 (L) are polynomials in the lag operator, so that current and lagged values of the exchange rate and foreign costs are allowed to affect import prices in order to capture slow adjustment. 1 Disaggregated monthly consumer price indices into 31 goods and services, for 8 regions in Egypt from July 2000 through July 2005 were used to isolate the impact of the exchange rate changes on consumer prices. 2 See for example Campa and Goldberg (2005) for a justification of this particular specification. Burstein, Eichenbaum and Rebelo (2005) document the importance of non-traded components of traded goods prices and their role in real exchange rate fluctuations. 3 These are the United States, Germany, Italy, Great Britain, and Japan. Saudi Arabia and France are among Egypt's top 5 sources of imports in 2000 but do not report monthly producer price indices. Page 22 15 Taking log differences of (1) and using (2) gives the growth rate of the consumer price as a function of the growth rate of the exchange rate: (3) ( ) irt t ir 2 t ir 1 ir 0 ir N irt ir irt u C ln ) L ( E ln ) L ( ) 1 ( P ln P ln D+ D ื b + D ื b + b ื a - + D ื a = D The effect of current and lagged changes in the exchange rate on consumer prices is given by ) L ( ) 1 ( ir 1 ir b ื a - , and this is the key parameter of interest for this section. It is important to note that the sensitivity of consumer prices to the exchange rate is likely to be substantially smaller than the sensitivity of border prices to the exchange rate. This is because consumer prices contain a substantial non-traded component, both in the form of non-traded items themselves, as well as distribution costs. Not have direct information on the size of these distribution margins is available in the case of Egypt, although in principle these can be extracted from input-output tables for Egypt. In industrial countries, these distribution margins are typically quite substantial, averaging 30-50 percent of the prices paid by consumers (Campa and Goldberg (2006). To address this problem an assumption is made that the growth rate of the non-traded component of the price of each item in each region consists of a common component and an idiosyncratic component that is orthogonal to movements in the exchange rate: (4) irt N rt N irt v P P +D = D ln ln Another assumption is the common component of non-traded goods prices can be approximated by a simple average of a few items in the consumer price index that appear to be primarily non-traded on a priori grounds. These are Domestic Services, and Restaurant and Hotel Services. 4 These two assumptions (of a common component in non-traded goods prices, approximated by these two particular prices) are clearly strong ones and open to debate. However, it is not clear what good alternatives might be available. In any case, denoting the growth rate of the simple average of these items in each region as N rt P ˆ ln D , the following empirical specification is obtained: (5) irt N rt ir 3 t ir 2 t ir 1 ir 0 irt e P ˆ ln C ln ) L ( E ln ) L ( P ln +D ื g + D ื g + D ื g + g = D where ir 0 ir ir 0 ) 1 ( bื a - = g is the intercept; ) L ( ) 1 ( ) L ( ir 1 ir ir 1 b ื a - = g captures the effect of the exchange rate on consumer prices; ) L ( ) 1 ( ) L ( ir 2 ir ir 2 b ื a - = g captures the effect of foreign costs on consumer prices; ir ir 3 a = g captures the contribution of changes in non-traded goods prices; and irt ir irt ir irt u ) 1 ( v e D ื a - + D ื a = is the error term. Equation (5) is estimated by ordinary least squares. In practice, all growth rates are measured as monthly observations on quarterly log differences, and it was allowed for 3, 6, and 9 month lags of these growth rates in the estimation. Since monthly data from July 2000 through July 2005 are available, this gives 60 monthly data points on which to estimate this specification for each item and region . 4 Other clearly largely-non-traded items are rent and education. However prices of these items are tightly controlled in Egypt and movements in them are unlikely to properly reflect movements in overall non-traded goods prices. Page 23 16 S TEP 2: E STIMATION OF THE W ELFARE E FFECT OF THE C HANGES IN P RICES I NDUCED B Y M OVEMENTS IN THE E XCHANGE R ATE The next step is to take the estimates of the changes in prices induced by movements in the exchange rate and calculate their welfare effects in terms of the compensating variation (a standard measure of welfare effects of price changes). To do this, we bring the estimated price changes due the devaluation for each of these 160 different price indices to the household survey for Egypt. This way we are able to empirically construct estimates of the compensating variation associated with these price changes for each household. In particular, we estimate how much higher (or lower) each household's total expenditure would have to be in order to attain the pre-devaluation level of utility at post-devaluation prices. 5 This compensating variation consists of two parts. The first is the change in the cost of households' initial consumption bundles as a result of devaluation-induced price changes. Considering only this direct effect would overstate the welfare effect of the price changes because it does not take into account how households change their spending patterns in response to price changes. If households can substitute away from goods that become relatively more expensive, then the direct effect of the price changes will exaggerate the welfare effects since it assumes no such substitution is possible. Therefore, there is a second component that captures changes in household behavior in response to these price changes. Estimating these substitution effects is therefore important, although substantially more involved for two reasons. The first is data limitations that would make results somewhat difficult to interpret. The second is that we need an estimate of substitution effects in response to depreciation-induced price changes, not one of substitution effects in response to overall price changes. The poverty line is compared to the counterfactual "post-depreciation" distribution of expenditure, to calculate the proportional change in the headcount between the two distributions. Thus, for example, for a choice of poverty line which delivers an initial headcount of 20 percent for Egypt as a whole, the effect of the depreciation was to raise the headcount by 23 percent, to 24.6 percent. Since we saw earlier that the average welfare effect of the depreciation was largest in Rural Upper Egypt, it is not surprising that the estimated poverty effects are largest here too. Finally, the role of substitution effects in response to the price changes induced by the depreciation is considered. Cohort techniques are used to estimate the change in household spending share using the 2000 and 2005 household surveys and from this the substitution effects. Cohorts are constructed based on four education categories, five age categories and seven regional categories, for a total of 140 cohorts. For each cohort, the average spending shares across the 20 expenditure items in the 2000 and 2005 surveys are calculated. Using the household-level variation within each cohort in the 2000 survey, cohort-specific income elasticities for each expenditure share are calculated. Finally, all these ingredients are combined with our estimates of the devaluation-induced price changes, to estimate the substitution effect. 5 See Friedman and Levinsohn (2002) for a similar exercise investigating the welfare effects of relative price changes following in Indonesia during the East Asian crisis of 1997. Page 24 17 A NNEX 3.2 M ETHODOLOGY OF S IMULATING THE P OVERTY P ATH C i,t is per capita consumption for sample household i in year t , where t=0 represents the survey year and g t is the consumption growth rate in year t. The basic form of the projections is to calculate per adult equivalent consumption recursively, C i,t = C i,t-1 ( 1+ g t ). In order to take into account changes in inequality, adjustment of consumption for each household within each sector year-by-year is made after the growth projection for each year. Take the percentage change in the Gini coefficient in year t as DG t The adjusted level of per capita consumption for household i is then C i,t , adj = C i,t -DG ( C t - C i,t ), Where C t is mean consumption in year t. This produces a proportional shift in Lorenz curve by adjusting consumption for each household relative to its deviation from the sector-specific mean. The Datt and Walker approach outlined above was designed to project changes in poverty based national growth data and a single household survey, in the absence of data from multiple household surveys. For the analysis constructed in this section, the goal is to understand changes in the distribution during the period between two household surveys. While in the Datt and Walker general case the distribution of consumption is known only at the beginning of the simulated period, for Egypt the full distribution is known both at the beginning and the end of the period 2000-2005. An extension of the Datt and Walker method was developed by Demombynes and Hoogeveen 6 , and used to force the distribution at the end of the simulation to closely match that of the survey data. Specifically, extend the notation from above to reference each household’s quantile, with n the number of fractions in which the distribution is broken down. For instance if n equals five, the distribution would broken down in quintiles. Let g t Qi is the consumption growth rate for the quantile of household i in year t. The simulation is carried out using these growth rates: ) 1 ( 1 , , i Q t g t i C t i C \0e \10 . Because these year-by-year sector-quantile growth rates are not directly observed, estimates must be used. What is observed (via survey-based estimates) is quantile’s growth in mean household consumption between the two surveys, denoted by G Q . Then 1 ) * 1 ( 1 * 1 1 545.4 545.4 11.3 4.6 45 : \10 525.1 525.1 11.3 4.6 45 : \0e 494.7 494.7 11.3 3.3 45 : ป 502.4 502.4 11.3 3.3 45 : ป 491.1 491.1 11.3 3.3 45 : ผ 510.1 510.1 11.3 3.3 45 : บ 465.2 465.2 11.3 3.3 45 : ซ 472.9 472.9 11.3 3.3 45 : ซ 461.6 461.6 11.3 3.3 45 : ฌ 480.6 480.6 11.3 3.3 45 : ช 474.2 474.2 11.3 4.6 45 : \0e 486.6 486.6 11.3 4.6 45 : \0e 465.5 465.5 11.3 4.6 45 : t g T G Q G Q t g i . An asterisk is used here to refer to overall growth versus growth in particular quantile. The effect is to scale the year-by-year growth rates from the national accounts for each quantile such that they cumulatively produce the growth observed in the survey for that quantile. When year-by- year growth rates calculated this way by sector and quantile are applied in the simulation, starting with the initial year survey data, they should produce a distribution in the final year which closely matches the final year survey data. For the analysis in this section, 50 quantiles are used, thus the simulated distribution in the final year is more closely matched with that in the final survey year data. Note that the initial and final years of the simulation will match the “true” distributions in survey data, and consequently the simulation’s cumulative change in mean consumption will match the change implied by the survey data. If cumulative growth in per capita consumption differs from cumulative consumption growth in the survey data, the simulated year-by-year changes in mean 6 Gabriel Demombynes and Johannes G. Hoogeveen, "Growth, Inequality and Simulated Poverty Paths for Tanzania, 1992-2002"WPS3432 . Page 25 18 consumption will not be equal to the year-by-year per capita changes. In Egypt, the overall per capita consumption growth (36 percent) is close to growth in consumption per capita in the survey (30 percent). As a result, the difference between the per capita consumption changes and simulated mean consumption changes in each year is not large. Page 26 19 ANNEX 4.1: E STIMATING H OUSEHOLD I NCOME P OVERTY To examine the interlinkage between labor market outcomes and household poverty, we need to estimate poverty levels for the ELMS 98 and ELMPS 06 household samples. Apart from looking at reported earnings and household asset index, household income poverty is estimated in this study using a two-stage estimation technique. This technique allows us to combine detailed income and expenditure information available from the HIECSs, with the rich labor market information available from the LMSs. The two stage approach will combine the HIECS 99 with the ELMPS 98, and HIECS 04 with the ELMPS 06 to estimate per capita consumption for the LMS samples. This will typically involve the following three steps: A. I DENTIFYING HOUSEHOLD CHARACTERISTICS AVAILABLE IN THE HIECS S AND THE ELMS S This stage involves comparing the HIECS and the ELMS questionnaires to identify common household variables found in the four datasets. This has not been a major constraint on the analysis, because a large set of common variables is available in all four datasets. In this paper, the choice of the final set of explanatory variables is based on a thorough review of the poverty literature and a careful investigation of the descriptive statistics of the common set of explanatory variables and their correlation with the poverty measures. B. E STIMATING PER CAPITA CONSUMPTION USING THE HIECS S DATA This stage is the first step of the two-step estimation approach. In this first-step, each of the two HIECS data is used to estimate per capita consumption as a function of the chosen common set of household characteristics. A log-linear function of per capita consumption of household i , y i , is estimated for each of the HIECS samples 7 : i i i X y e b + ข = ln where X i is a vector of cluster-level characteristics of household i ; and e i is a disturbance term that is distributed as N(0, s 2 ). Of course, some of the explanatory variables selected in the first stage are endogenous, which would bias the estimation results. For instance, the ownership of durables are particularly among the set of endogenous variables, since it is closely determined by the household living standard and thus by the poverty status (Astrup and Dessus 2001). However, as discussed in Minot (2000), the possible endogeneity of some of the explanatory variables is less of a concern in the current analysis since the main objective here is to predict the level of poverty (or ln y i ), rather than to study the determinants of poverty or to assess the impact of each explanatory variable. C. P REDICTING PER CAPITA CONSUMPTION FOR THE ELMS S SAMPLES In this stage, the regression models developed in the previous step and the ELMSs data are used to predict per capita consumption for each of the two rounds of ELMSs. 7 This paper uses consumption rather than income to measure household welfare. Consumption is often preferred over income when measuring welfare, since consumption data is likely to be subject to less fluctuation over time and to fewer measurement errors (see Deaton 1997). Page 27 20 A NNEX T ABLES Daily calories intake Quantity in gms of total calories Cereals and starches Meat and poultry Fish milk milk products Eggs oil butter fruits vegetables Pulses sugar others Tea coffee drinks soft drinks Total Table A.1.2: Quantities and Calories Generated by the Reference Food B undle Table A.1.1: Daily Caloric Requirements by Age, Sex and Location Age Group Male Female Male Female Urban Rural Page 28 21 Table A.1.3: Cost of 100 Calories by Region Region Cost of calories PT Metropolitan Lower Urban Lower Rural Upper Urban Upper Rural Border Urban Border Uural 1995/96 1999/2000 2004/2005 Number of Households Number of Individuals Number of Households Number of Individuals Number of Households Number of Individuals 6622 28911 28754 125287 21743 88843 8183 45028 19195 100830 25352 118588 14805 73939 47949 226117 47095 207431 Table A.1.4: Sample Size of 1995/96, 1999/00 and 2004/05 Surveys Page 29 22 Table A.1.5: Regression Results (Dependent Variable ln Household Expenditure), 1995/96, Metropolitan Coefficient Std Err t Prob t Cairo Alexandria Suez Age of the head of household Head has above secondary education head is self employed head is illiterate head has other than secondary education head is male head has university degree head is government employee head works in agriculture activity Has private kitchen ln household size squared ln household size has private bathroom Share of adult females Share of adult males Has no sewerage system share of children of age and under share of employed persons share of employers in agricultural activities share of employers in non agricultural activities share of self employed persons in non agricultural activities share of employed persons in private sector share of illiterates share of out of labor force persons share of unemployed persons share of university graduates has tapped water Adjusted R Number of households Page 30 23 Table A.1.6: Regression Results (Dependent Variable ln Household Expenditure), 1995/96, Lower Urban Coefficient Std Err t Prob t has electricity Sharkia Qualiobia Garbeyya Menoufia Beheira Ismailia head has no basic education head is employed head is literate head has a degree in education head is male head has university degree head is employed in private sector has kitchen ln household size share of adult males share of children share of government employees share of employers in agriculture share of wage workers not in agricultural activities share of employers not in agricultural activities share of out of labor force persons has tapped water Adjusted R Number of households Page 31 24 Table A.1.7: Regression Results (Dependent Variable ln Household Expenditure), 1995/96, Lower Rural Coefficient Std Err t Prob t has electricity Sharkia Qualiobia Garbeyya Beheira Ismailia ln household size age of head head has no basic education head is literate head has a degree in education head has no secondary education degree head is employer in agricultural activities head is government employee has kitchen ln household size has access to public water network Share of adult females Share of adult males share of children of age and under share of government employees share of wage workers in agricultural activities share of employers in agricultural activities share of self employed in agricultural activities share of unpaid workers in agricultural activities share of employers in non agricultural activities share of employed persons in private sector share of out of labor force persons has tapped water Adjusted R Number of households Page 32 25 Table A.1.8: Regression Results (Dependent Variable ln Household Expenditure), 1995/96, Upper Urban Coefficient Std Err t Prob t squared age of head age of the head head has no above secondary education head is illiterate head has secondary degree head has university degree head is not employer in agricultural activities head is not self employed in agricultural activities head is not government employee head is not employer in non agricultural activities head is not employed in private sector ln household size squared household size does not have private bath Share of adult females Share of adult males share of children of age and under share of wage workers in agricultural activities share of self employed in agricultural activities share of employers in non agricultural activities share of illiterates share of unemployed persons shares of university graduates has tapped water Adjusted R Number of households Page 33 26 Table A.1.9: Regression Results (Dependent Variable ln Household Expenditure), 1995/96, Upper Rural Coefficient Std Err t Prob t head is wage worker head is employer head is literate head is male head has university degree head is not employer in agricultural activities head is self employed in agricultural activities head is not government employee squared ln household size ln household size has private bathroom share of adult males has sewerage system share of children share of children of age and under share of employed persons share of wage workers in agricultural activities share of employers in non agricultural activities share of self employed in non agricultural activities share of wage workers in non agricultural activities share of illiterates share of out of labor force persons share of unemployed persons has tapped water mean years of schooling Adjusted R Number of households Page 34 27 Table A.1.10: Regression Results (Dependent Variable ln Household Expenditure), 1995/96, Border Urban Coefficient Std Err t Prob t head has above secondary degree head is unemployed head has a degree in education ln household size share of employed persons share of employers in agricultural activities mean years of schooling Adjusted R Number of households Table A.1.11: Regression Results (Dependent Variable ln Household Expenditure), 1995/96, Border Rural Coefficient Std Err t Prob t squared age of head head is unemployed head is out of labor force head is literate head has no secondary degree head is not government employee ln household size has access to public water network has private kichen has access to sewerage system Adjusted R Number of households Page 35 28 Table A.1.12 Distribution of Poorest 50, 100 and 200 Sub-Districts by Governorate, 1996 Sub Districts Counts Sub Districts Counts Sub Districts Counts Upper Egypt Border South Sinai North Sinai Matrouh El Wadi El Gidid Red Sea Aswan Qena Sohag Lower Egypt Metropolitan Menia Assiut Giza Beni Suef Fayoum Gharbia Menufia Beheira Ismailia Sharkia Qaliubia Kafr El Sheikh Suez Damietta Dakahlia Cairo Alexandria Port Said Poorest Sub Districts Poorest Sub Districts Poorest Sub Districts Incidence Incidence Incidence Page 36 29 Table A.1.13 Distribution of Poorest 50, 100 and 200 Sub-Districts by Governorate, 2006 Incidents Incidents Incidents Total Sub Districts All Egypt Luxor Aswan Qena Sohag Assiut Menia Fayoum Beni Suef Giza Ismailia Beheira Menufia Gharbia Kafr El Sheikh Qaliubia Sharkia Dakahlia Damietta Suez Port Said Alexandria Cairo Sub Districts Counts Sub Districts Counts Sub Districts Counts Poorest Sub Districts Poorest Sub Districts Poorest Sub Districts Page 37 30 Table A.1.14 Distribution of Poorest 100, 500 and 1000 Villages by Governorate, 1996 Incidence% Villages counts Incidence% Villages counts Incidence% Villages counts Damietta Dakahlia Sharkia Qaliubia Kafr El-Sheikh Gharbia Menufia Beheira 0.5 1 2.9 6 6.8 14 Ismailia Giza 0.6 2 10.1 16 27.7 44 Beni-Suef 12.7 28 45.5 100 76.8 169 Fayoum 8.2 13 37.7 60 69.2 110 Menia 4.3 15 39.3 136 78.3 271 Assiut 5.6 13 25.6 60 56.8 133 Sohag 4.4 12 24.1 65 54.1 146 Qena 8.1 16 25.9 51 51.8 102 Aswan 3.4 3 6.8 6 Red Sea El Wadi El- Gedid Matrouh 4.0 3 5.3 4 North Sinai South Sinai 1.2 1 Overall Egypt 3.0 100 15.0 500 29.9 1000 Poorest 100 Villages Poorest 500 Villages Poorest 1000 Villages Page 38 31 Table A.1.15 Distribution of Poorest 100, 500 and 1000 Villages by Governorate, 2006 Incidents villages Counts Incidents villages Counts Incidents villages Counts Damietta Dakhlia Sharkia Qualyoubia Kafr Elsheikh Garbia Menofia Behera Ismailia Giza Beni Suef Fayoum M enia Assuit Sohag Qena Aswan Luxor All Egypt Total number of villages Poorest villages Poorest villages Poorest villages Page 39 32 Table A.2.1 (a): Poverty Measurements by Educational Attainment of Individuals, 2004-05 (percent) R egion Illiterate Read & w rite B asic Secondary Diploma University . Post University . T otal Metropolitan \03 \03 \03 \03 \03 \03 \03 P0 10.85 8.50 7.36 4.10 2.73 0.85 0.00 5.95 P1 1.76 1.29 1.14 0.52 0.36 0.09 0.00 0.89 P2 0.45 0.32 0.28 0.11 0.09 0.02 0.00 0.22 no individuals 5133 5065 6734 8175 1304 5311 200 31922 Lower Egypt Urban P0 15.00 10.94 10.53 7.63 6.98 3.05 1.62 9.60 P1 2.55 1.75 1.55 1.11 1.09 0.43 0.11 1.50 P2 0.65 0.44 0.36 0.27 0.26 0.10 0.01 0.37 no individuals 3998 3398 3881 5575 747 2497 60 20156 Lower Egypt Rural P0 21.69 17.57 18.41 14.33 9.27 6.99 0.00 17.58 P1 3.24 2.39 2.57 2.06 1.41 0.84 0.00 2.52 P2 0.75 0.51 0.55 0.47 0.32 0.17 0.00 0.57 no individuals 15856 8868 9866 11864 942 2480 32 49908 Upper Egypt \03\03 U rban P0 29.53 20.19 20.87 14.71 7.90 4.99 0.00 18.64 P1 6.41 4.42 4.34 2.69 1.30 0.88 0.00 3.84 P2 2.00 1.40 1.32 0.75 0.34 0.24 0.00 1.17 no individuals 4300 3058 4151 4951 739 2299 110 19608 Upper Egypt \03\03 R ural P0 43.03 40.93 41.52 33.15 26.32 21.75 0.00 39.87 P1 9.20 8.33 8.69 6.67 5.18 4.06 0.00 8.33 P2 2.82 2.46 2.63 1.98 1.52 1.15 0.00 2.51 no individuals 16779 6274 7740 6515 543 1148 11 39010 All Egypt P0 28.46 20.76 20.93 14.28 8.63 4.74 0.23 19.75 P1 5.51 3.73 3.83 2.45 1.46 0.74 0.02 3.64 P2 1.59 1.03 1.06 0.66 0.39 0.19 0.00 1.02 no individuals 46459 26977 32800 37554 4368 13916 416 162490 Page 40 33 Table A.2.1 (b): Poverty Measurements by Educational Attainment of Individuals 2004-05 (percent) Regi o n Illiter a t e Ca n R W doe s not hol d a De g r ee Bel o w Aver a ge De g r ee Pr i ma r y Pr e pa r at o r y Aver a ge De g r ee Secon d a r y De g r ee or equ i v al e nt Ab o ve Aver a ge De g r ee bu t be l ow Un i v er s i ty De g r ee Un i v er s i ty De g r ee A b ove Un i v er s i ty De g r ee ma s t er s ph D To t al Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Up p e r Ru r a l Over al l E gyp t Ur b a n Over al l E gyp t Ru r a l To t al Me t rop ol i tan Lower Ur b a n Lower Ru r a l Up p e r Ur b a n Page 41 34 Table A.2.2 (a): Educational Status of Individuals by Region by Poverty Status, 2004- 05 (percent) Illiterate Read write Primary Secondary Diploma University Post University Total Individuals Mertopolitan Non poor Poor All Lower Urban Non poor Poor All Lower Rural Non poor Poor All Upper Urban Non poor Poor All Upper Rural Non poor Poor All All Egypt Non poor Poor All Page 42 35 Table A.2.2 (b): Educational Status of Individuals by Region by Poverty Status 2004-05 (percent) Re g i o n Ill i t e r a t e Ca n R W do e s no t ho l d a De g r e e Be l o w Av e r a g e De g r e e Pr i ma r y Pr e p a r at o r y Av e r a g e De g r e e Se c o n d a r y De g r e e or eq u i v a l e n t Ab o v e Av e r a g e De g r e e bu t be l o w Un i v e r s i t y De g r e e Un i v e r s i t y De g r e e Ab o v e Un i v e r s i t y De g r e e ma s t e r s ph D To t a l Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Me t r o p o l i t a n Lo w e r Ur b a n Lo w e r Ru r a l Up p e r Ur b a n Up p e r Ru r a l Ov e r a l l Eg y p t Ur b a n Ov e r a l l Eg y p t Ru r a l To t a l Page 43 36 Table A.2.3 (a): Poverty Measurements by Employment Status of individuals, 2004-05 (percent) Wage Workers Employer Self employed Unpaid Worker Unemployed Out of labor force Out of human force Total M etropolitan P0 4.98 1.97 6.89 12.68 10.26 6.39 3.97 5.67 P1 0.70 0.30 1.25 3.31 1.85 0.96 0.57 0.85 P2 0.16 0.08 0.32 1.19 0.57 0.23 0.13 0.21 No. Individuals 9850 999 1207 119 865 19719 5883 38642 Lower Urban P0 7.79 8.11 8.61 21.81 13.32 9.79 6.07 9.00 P1 1.22 1.08 1.47 3.17 2.22 1.53 0.81 1.38 P2 0.31 0.22 0.40 0.72 0.58 0.38 0.17 0.34 No. Individuals 5741 1088 1358 602 788 11543 3991 25111 Lower Rural P0 16.73 13.61 13.15 22.93 24.12 18.13 12.33 16.66 P1 2.45 1.85 1.82 3.33 4.20 2.56 1.70 2.38 P2 0.56 0.40 0.40 0.78 1.11 0.56 0.38 0.53 No. Individuals 11661 4573 5613 5321 1318 24538 10635 63659 Upper Urban P0 15.71 12.53 22.53 28.90 26.55 20.08 15.42 18.60 P1 3.16 2.50 4.27 5.52 5.37 4.27 3.06 3.83 P2 0.95 0.74 1.17 1.52 1.59 1.32 0.89 1.15 No. Individuals 5380 815 1309 437 712 11946 3937 24536 Upper Rural P0 42.30 28.67 33.54 35.99 42.87 43.82 33.29 39.06 P1 9.26 5.34 6.45 6.82 9.77 9.32 6.53 8.07 P2 2.90 1.49 1.83 1.94 3.05 2.84 1.90 2.42 No. Individuals 7403 3600 4166 4530 688 22113 9995 52495 All Egypt P0 17.03 16.82 19.30 28.36 22.57 21.01 16.55 19.56 P1 3.18 2.82 3.40 4.85 4.41 3.96 2.96 3.60 P2 0.91 0.73 0.91 1.29 1.28 1.13 0.81 1.01 No. Individuals 40586 11166 13767 11083 4419 91056 34955 207032 Page 44 37 Table A.2.3 (b): Poverty Measurements by Employment Status of Individuals 2004-05 (percent) Region Wage Earner S elf E mployed h iring o thers S elf E mployed w orking a lone U npaid W orker Unemployed Total Extreme Poor Poor N ear P oor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor P oor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Total Overall Egypt Urban M etropolitan Lower Urban Lower Rural Upper Urban Upper Rural Overall Egypt Rural Page 45 38 Table A.2.4 (a): Employment Status of Individuals by Region by Poverty Status 2004-05 (percent) Wage Workers Employer Self employed Unpaid Worker Unemployed Out of labor force Out of human force Total Metropolitan Non poor Poor Total Lower Urban Non poor Poor Total Lower Rural Non poor Poor Total Upper Urban Non poor Poor Total Upper Rural Non poor Poor Total All Egypt Non poor Poor Total Page 46 39 Table A.2.4 (b): Employment Status of Individuals by Region by Poverty Status, 2004- 05 (percent) Re g i on Wag e Ea r ne r Sel f Em p l oy e d hi r i ng ot h e r s sel f Em p l oy e d wo r k i ng al o ne Un p a i d Wo r ke r Un e mp l oy e d Ou t of Lab o r Fo r ce Ou t of Hu m a n Fo r ce to t a l Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Up p e r Ru r a l Ov e r al l Eg y pt Ur b a n Ov e r al l Eg y pt Ru r a l To t al Me t ro p o l i t an Lo wer Ur b a n Lo wer Ru r a l Up p e r Ur b a n Page 47 40 TableA.2.5 (a): Employment Status of Labor Force by Region by Poverty Status 2004- 05 (percent) Wage Worker Employer self employed Unpaid worker Unemployed Total Labor Force Metropolitan Non poor Poor Total Lower Urban Non poor Poor Total Lower Rural Non poor Poor Total Upper Urban Non poor Poor Total Upper Rural Non poor Poor Total All Egypt Non poor Poor Total Page 48 41 Table A.2.5 (b): Employment Status of Labor Force by Region by Poverty Status 2004-05 (percent) Re g i on Wa g e Ea r ne r Se l f Em p l oy e d hi r i ng ot h e r s se l f Em p l oy e d wo r ki n g al o n e Un p a i d Wo r k e r Un e mp l oy e dT o ta l Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Me t r op o l i t an Lo we r Ur b a n Lo we r Ru r a l Up p e r Ur b a n Up p e r Ru r a l Ov e r a l l Eg y p t Ur b a n Ov e r a l l Eg y p t Ru r a l To t al Page 49 42 Table A.2.6 (a): Poverty Measurements by Sector of Employment of Individuals 2004- 05 (percent) Government Investment Public Private Co operative NGO Foreign Outside Establishments Total Metropolitan P P P No Individuals Lower Urban P P P No Individuals Lower Rural P P P No Individuals Upper Urban P P P No Individuals Upper Rural P P P No Individuals All Egypt P P P No Individuals Page 50 43 Table A.2.6 (b): Poverty Risk by Sector of Employment of Individuals 2005 (percent) Re g i on Go v e r nm e n t In v e s t me n t Pu b l i c Pr i va t e Co o p e r a t i ve NG O Fo r e i gn JV Ou t si d e e st a b l i s h me n t s To t al Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Up p e r Ru r a l Ov e r a l l Eg y p t Ur b a n Ov e r a l l Eg y p t Ru r a l To t al M et r o p o l i t an Lo w e r Ur b a n Lo w e r Ru r a l Up p e r Ur b a n Page 51 44 Table A.2.7 (a): Sector of Employment of Labor Force by Region by Poverty Status 2004-05 (percent) Government Investment Public Private Co operative NGO Foreign Outside Establishments Total Metropolitan Non poor Poor Total Lower Urban Non poor Poor Total Lower Rural Non poor Poor Total Upper Urban Non poor Poor Total Upper Rural Non poor Poor Total All Egypt Non poor Poor Total Page 52 45 Table A.2.7 (b): Sector of Employment of Labor Force by Region by Poverty Status 2004-05 (percent) Re g i o n Go v e r n m e n t In v e s t me n t Pu b l i c Pr i v a t e Co o p e r a t i v e NG O Fo r e i g n JV Ou t si d e es t a b l i s h m e n t s To t a l Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Me t r o p o l i t a n Lo w e r Ur b a n Lo w e r Ru r a l Up p e r Ur b a n Up p e r Ru r a l Ov e r a l l Eg y p t Ur b a n Ov e r a l l Eg y p t Ru r a l To t a l Page 53 46 Table A.2.8 (a): Poverty Measurements by Economic Activity of Individuals 2004-05 (percent) Agriculture Mining Manufacturing Electricity Construction Trade Transportation Finance Services Total Metropolitan P P P No Individuals Lower Rural P P P No Individuals Upper Urban P P P No Individuals Upper Rural P P P No Individuals All Egypt P P P No Individuals P P P No Individuals Page 54 47 Table A.2.8 (b): Poverty Measurements by Economic Activity of Individuals, 2004-05 (percent) R eg i o nA g ri c u l t ur eM i ni n gM a nu f a c t o r i ng El e c t r i c i t yC o n s t r u c t i o n Tr a d e Ho t e l a nd Re s t a u r a n t T ra n s p o r t F in a n c e a nd Re a l Es t a t e Pu b l i c an d F am i l y Se r v i c e s T ot a l Extreme Poor Poor Near Poor Better Of f Extreme Poor Poor Near Poor Better Of f Extreme Poor Poor Near Poor Better Of f Extreme Poor Poor Near Poor Better Of f Extreme Poor Poor Near Poor Better Of f Extreme Poor Poor Near Poor Better Of f Extreme Poor Poor Near Poor Better Of f Extreme Poor Poor Near Poor Better Of f Up p e r Ru r a l Ov e r a l l Eg y p t Ur b a n Ov e r a l l Eg y p t Ru r a l To t a l Me t r o p o l i t a n L ow e r U rb a n Lo w e r Ru r a l Up p e r Ur b a n Page 55 48 Table A.2.9 (a): Economic Activity of Labor Force by Region by Poverty Status 2004-05 (percent) Agriculture Mining Manufacturing Electricity Construction Trade Transport ation - Finance Services Total Metropolitan Non- poor 1.73 0.49 19.11 1.41 9.54 21.44 12.18 3.96 30.15 11640 Poor 2.58 24.96 1.13 16.59 28.34 8.21 1.93 16.26 621 All 1.77 0.46 19.40 1.39 9.89 21.79 11.98 3.86 29.44 12261 Lower Urban Non- poor 16.88 0.15 15.67 1.36 7.07 17.82 7.62 2.29 31.14 8034 Poor 24.81 0.13 19.47 0.51 11.07 22.52 5.85 0.89 14.76 786 All 17.59 0.15 16.01 1.28 7.43 18.24 7.46 2.17 29.68 8820 Lower Rural Non- poor 53.05 0.11 8.49 0.70 5.03 8.26 4.51 0.82 19.01 22669 Poor 62.95 0.02 8.23 0.29 5.91 7.19 3.75 0.26 11.40 4534 All 54.70 0.10 8.45 0.63 5.18 8.08 4.38 0.73 17.74 27203 Upper Urban Non- poor 11.36 0.33 12.33 1.00 7.53 21.48 9.23 3.06 33.70 6612 Poor 24.20 0.15 12.94 0.58 13.37 20.13 7.34 1.09 20.20 1376 All 13.57 0.30 12.43 0.93 8.54 21.24 8.90 2.72 31.37 7988 Upper Rural Non- poor 61.35 0.06 5.87 0.63 6.08 7.97 3.92 0.70 13.42 12522 Poor 63.99 0.11 4.70 0.38 9.81 7.70 2.60 0.18 10.54 7199 All 62.31 0.08 5.44 0.54 7.44 7.87 3.44 0.51 12.37 19721 All Egypt Non- poor 35.63 0.21 11.23 0.95 6.62 13.35 6.79 1.83 23.40 62196 Poor 55.34 0.09 8.20 0.41 9.22 10.35 3.80 0.40 12.18 14624 All 39.38 0.18 10.65 0.85 7.11 12.78 6.22 1.56 21.27 76820 Page 56 49 Table A.2.9 (b): Economic Activity of Labor Force by Region by Poverty Status 2004-05 (percent) Re g i on A g r i c u l t ur eM i ni n gM a nu f a c t or i ng El e c t r i c i t yC o ns t r uc t i on Tr a d e Ho t e l an d Re s t a u r a n t Tr a n s p o r t Fi n a n c e an d Re a l Es t a t e Pu b l i c an d Fa m i l y Se r v i c e s To t a l Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Me t r o p o l i t a n Lo w e r Ur b a n Lo w e r Ru r a l Up p e r Ur b a n Up p e r Ru r a l Ov e r a l l Eg y p t Ur b a n Ov e r a l l Eg y p t Ru r a l To t a l Page 57 50 Table A.2.10 (a): Poverty Measurements by Employment Type of Individuals 2005 (percent) Permanent Temporary Seasonal Occasional Total M etropolitan P P P No Individuals Lower Urban P P P No Individuals Lower Rural P P P No Individuals Upper Urban P P P No Individuals Upper Rural P P P No Individuals All Egypt P P P No Individuals Page 58 51 Table A.2.10 (b): Poverty Risk by Employment Type of Individuals and by Region, 2005 (percent) Regi o n Pe r ma n e nt Te m p or a r ySeas on a l Occas i on a lT o tal Me t rop ol i tan Extreme Poor Poor Near Poor Better Off Lower Ur b a n Extreme Poor Poor Near Poor Better Off Lower Ru r a l Extreme Poor Poor Near Poor Better Off Up p e r Ur b a n Extreme Poor Poor Near Poor Better Off Up p e r Ru r a l Extreme Poor Poor Near Poor Better Off Over al l E gyp t Ur b a n Extreme Poor Poor Near Poor Better Off Over al l E gyp t Ru r a l Extreme Poor Poor Near Poor Better Off To t al Extreme Poor Poor Near Poor Better Off Page 59 52 Table A.2.11 (a): Type of Employment of Individuals in Labor Force by Region by Poverty Status 2004-05 (percent) Permanent Temporary Seasonal Occasional Total M etropolitan Non poor Poor Total Lower Urban Non poor Poor Total Lower Rural Non poor Poor Total Upper Urban Non poor Poor Total Upper Rural Non poor Poor Total All Egypt Non poor Poor Total Page 60 53 Table A.2.11 (b): Type of Employment of Individuals in Labor Force by Region by Poverty Status 2004-05 (percent) Regi o n Pe r ma n e nt Te m p or a r ySeas on a l Occas i on a lT o tal Me t rop ol i tan Extreme Poor Poor Near Poor Better Off Lower Ur b a n Extreme Poor Poor Near Poor Better Off Lower Ru r a l Extreme Poor Poor Near Poor Better Off Up p e r Ur b a n Extreme Poor Poor Near Poor Better Off Up p e r Ru r a l Extreme Poor Poor Near Poor Better Off Over al l E gyp t Ur b a n Extreme Poor Poor Near Poor Better Off Over al l E gyp t Ru r a l Extreme Poor Poor Near Poor Better Off To t al Extreme Poor Poor Near Poor Better Off Page 61 54 Table A.2.12 (a): Poverty Measurements by Household Size 2004-05 (percent) One person Two persons Three persons Four persons persons or persons persons or more Total Metropolitan P P P Lower Urban P P P Lower Rural P P P Upper Urban P P P Upper Rural P P P All Egypt P P P Page 62 55 Table A.2.12 (b): Poverty Measurements by Household Size 2004-05 (percent) Region person persons persons persons persons or persons persons or more Extreme Poor P oor Near Poor B etter O ff Extreme Poor Poor Near Poor B etter O ff E xtreme P oor P oor Near Poor B etter O ff Extreme Poor Poor N ear P oor Better Off Extreme Poor P oor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Metropolitan Lower Urban Lower Rural Total U pper U rban Upper Rural Overall Egypt Urban Overall Egypt Rural Page 63 56 Table A.2.13 (a): Distribution of Individuals by Household Size, by Region and by Poverty Status 2004-05 (percent) One person Two persons Three persons Four persons persons or persons persons or more Total Metropolitan Non poor Poor Total Lower Urban Non poor Poor Total Lower Rural Non poor Poor Total Upper Urban Non poor Poor Total Upper Rural Non poor Poor Total All Egypt Non poor Poor Total Page 64 57 Table A.2.13 (b): Distribution of Individuals by Household Size, by Region and by Poverty Status 2004-05 (percent) Re g i on pe r s o n pe r s o n s pe r s o n s pe r s o n s pe r s o n s or pe r s o n s pe r s o n s or mo r e Extreme Poor Poor Near Poor Better Of f Extreme Poor Poor Near Poor Better Of f Extreme Poor Poor Near Poor Better Of f Extreme Poor Poor Near Poor Better Of f Extreme Poor Poor Near Poor Better Of f Extreme Poor Poor Near Poor Better Of f Extreme Poor Poor Near Poor Better Of f Extreme Poor Poor Near Poor Better Of f Lo w e r Ru r a l Up p e r Ur b a n Up p e r Ru r a l Ov e r a l l Eg y p t Ur b a n Ov e r a l l Eg y p t Ru r a l To t a l Me t r o p o l i t a n Lo w e r Ur b a n Page 65 58 Table A.2.13 (c) Poverty Risk of Households by Number of Children, by Region and Poverty Status, 2005 (percent) Reg i o n h ou s e hol d s with no ch i l de r n h ou s e hol d s wi t h on e ch i l d h ou s e hol d s wi t h two ch i l dr e n h ou s e hol d s wi t h th r e e ch i l dr e n hous e hol d s with mo r e th a n th r e e ch i l dr e n To t al Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Up p e r Ru r a l Over al l Eg y p t Ur b a n Over al l Eg y p t Ru r a l To t al Me t ro p o l i tan Lo wer Ur b a n Lo wer Ru r a l Up p e r Ur b a n Page 66 59 Table A.2.14: Average Household Size by Poverty Status for 2004-05 and 1999-00 C hange Non Poor Poor All Non Poor Poor All Non Poor Poor All Metropolitan Lower Urban Lower Rural Upper Urban Upper Rural All Egypt Table A.2.15 (a) Demographic Characteristics by Poverty Status and Region 2004-05 Average Number of children Average Number of adult males Average Number of adult females Average Number of elderly Average Household Size Metropolitan Non Poor Poor Total Lower Urban Non Poor Poor Total Lower Rural Non Poor Poor Total Upper Urban Non Poor Poor Total Upper Rural Non Poor Poor Total All Egypt Non Poor Poor Total Page 67 60 Table A.2.15 (b): Demographic Characteristics by Poverty Status and Region 2004-05 Reg i o nNu m b er of ch i l dr e n Nu m b er of ad u l t ma l es Nu m b er of ad u l t fem a l es Nu m b er of el d e r l y Hou s eh ol d Size Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor B etter O ff Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Up p e r Ru r a l Over al l E gyp t Ur b a n Over al l E gyp t Ru r a l To t al Me t rop ol i tan Lower Ur b a n Lo wer Ru r a l Up p e r Ur b a n Page 68 61 Table A.2.16: Poverty Measurements by Household Structure and Gender of Household Head, 2004-05 (percent) M ar r i ed w ith n o c hi l dr e n M ar r i ed w ith c hi l dr e n Ma r ri e d w ith mo r e t ha n th r e e c hi l dr e n W id o wed w ith n o c hi l dr e n W id o wed w ith c hi l dr e n Wid o wed w ith mo r e t ha n th r e e c hi l dr e n Nev er m ar r i ed O th e r s T ot a l Male Headed Households in Urban Areas P P P Female Headed Households in Urban Areas P P P All Households in Urban Areas P P P Male Headed Households in Rural Areas P P P Female Headed Households in Rural Areas P P P All Households in Rural Areas P P P Male Headed Households in All Egypt P P P Female Headed Households in All Egypt P P P All Households in All Egypt P P P Page 69 62 Table A.2.17: Distribution of Individuals by Household Structure, by Gender of Household Head and by Poverty Status, 2004-05 (percent) ma r ri e d with no ch i l dr e n ma r ri e d with ch i l dr e n ma r ri e d with mo r e th a n th r e e ch i l dr e n wid owed with no ch i l dr e n wid owed with ch i l dr e n wid owed with mo r e th a n th r e e ch i l dr e n ne v e r ma r ri e d ot h e r s To t al Male Headed Households in Urban Areas Non poor Poor Total Female Headed Households in Urban Areas Non poor Poor Total All Households in Urban Areas Non poor Poor Total Male Headed Households in Rural Areas Non poor Poor Total Female Headed Households in Rural Areas Non poor Poor Total All Households in Rural Areas Non poor Poor Total Male Headed Households in All Egypt Non poor Poor Total Female Headed Households in All Egypt Non poor Poor Total All Households in All Egypt Non poor Poor Total Page 70 63 Table A.2.18 (a): Poverty Measurements by Gender of Household Head, 2005(percent) Ma l e Head e d Hou s eh ol d s Fem a l e Head e d Hou s eh ol d s To t al Metropolitan P P P Lower Urban P P P Lower Rural P P P Upper Urban P P P Upper Rural P P P All Egypt P P P Page 71 64 Table A.2.18 (b): Poverty Risk by Gender of Household Head, 2005 (percent) Regi o n Ma l e Fem a l eT o tal Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Over al l Egyp t Ru r a l To t al M et r op ol i tan Lower Ur b a n Lower Ru r a l Up p e r Ur b a n Up p e r Ru r a l Over al l Egyp t Ur b a n Page 72 65 Table A.2.19 (a): Distribution of Individuals by Gender of Household Head, by Region and Poverty Status, 2005 Male Headed Households Female Headed Households Total Metropolitan Non poor Poor Total Lower Urban Non poor Poor Total Lower Rural Non poor Poor Total Upper Urban Non poor Poor Total Upper Rural Non poor Poor Total Upper Rural Non poor Poor Total Page 73 66 Table A.2.19 (b): Distribution of Individuals by Gender of Household Head, by Region and Poverty Status, 2005 Regi o n M al e F em a l eT o tal Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Extreme Poor Poor Near Poor Better Off Lower Ru r a l Up p e r Ur b a n Up p e r Ru r a l Over al l Egyp t Ur b a n Over al l Egyp t Ru r a l To t al Me t rop ol i tan Lower Ur b a n Page 74 67 Table A.2.20: Illiteracy Rate among Children of Age 12-15 Years Old by Poverty Status and Region 2004-05 Boys Girls Total Male headed households Female headed households Metropolitan Non Poor Poor Total Lower Urban Non Poor Poor Total Lower Rural Non Poor Poor Total Upper Urban Non Poor Poor Total Upper Rural Non Poor Poor Total All Egypt Non Poor Poor Total Page 75 68 Table A.2.21: Percentage of Working Children Aged 6-15 Years by Poverty Status and Gender, 2004-05 Boys Girls Total M ale h eaded households F emale h eaded households Metropolitan Non Poor Poor Total Lower Urban Non Poor Poor Total Lower Rural Non Poor Poor Total Upper Urban Non Poor Poor Total Upper Rural Non Poor Poor Total All Egypt Non Poor Poor Total Page 76 69 Table A.2.22: Net Enrolment Rate in Basic Education by Poverty Status and Gender 2004-05 (percent) Boys Girls Total Male headed households Female headed households Metropolitan Non Poor Poor Total Lower Urban Non Poor Poor Total Lower Rural Non Poor Poor Total Upper Urban Non Poor Poor Total Upper Rural Non Poor Poor Total All Egypt Non Poor Poor Total Page 77 70 Table A.2.23: Shares of Different Income Sources by Poverty Status and Gender of Household Head 2004-05 wag es an d sal a r i es ag r i cu l t ur a l pr o j ec t s no n ag r i cu l t ur a l pr o j ec t s fi n a nc i al as s et s no n fi n a nc i al as s et s im p u t ed ren t fo r ho us e s tran s f er s To t al Female Headed households in Urban Areas Non Poor Poor Total Male Headed households in Urban Areas Non Poor Poor T otal All households in Urban Areas Non Poor Poor Total Female Headed households in Rural Areas Non Poor Poor Total Male Headed households in Rural Areas Non Poor Poor Total All households in Rural Areas Non Poor Poor Total Female Headed households in All Egypt Non Poor Poor Total Male Headed households in All Egypt Non Poor Poor Total All Households in All Egypt Non Poor Poor Total Page 78 71 Table A.2.24: Percentage Shares of Different Types of Transfers , Out of Total Income, by Poverty Status and Gender of Household Head 2004-05 Government pension Social insurance Sadat pensions Pensions from unions Social guaranty Transfers from outside the country Transfers within the country Other periodical revenues donations Total transfers Female Headed households in Urban Areas Non Poor Poor Total Male Headed households in Urban Areas Non Poor Poor Total All households in Urban Areas Non Poor Poor Total Female Headed households in Rural Areas Non Poor Poor Total Male Headed households in Rural Areas Non Poor Poor Total All households in Rural Areas Non Poor Poor Total Female Headed households in All Egypt Non Poor Poor Total Male Headed households in All Egypt Non Poor Poor Total All Households in All Egypt Non Poor Poor Total Page 79 72 Table A.2.25: Percentage of Households with Public Amenities Characteristics by Poverty Status 2004-05 urban Rural n on p oor p oor T otal n on p oor p oor T otal public network tap inside dwelling connected to sewerage network have electricity have private kitchen have proper means of garbage collection abnormal water color abnormal water smell there is an obstruction or superabundance in the system No health office no public health center Page 80 73 Table A.2.26: Percentage of Households by Ownership of Durable Goods and by Poverty Status 2004-05 Urban Rural All Egypt N on poor Poor Total N on poor Poor Total N on poor Poor Total private car bicycles motocycle telephone cellular phone Internet Refrigerator Deep Freezer Gas bottles Electricity microwave normal electric washing machine Automatic washing machine electric dish machine Gas electric water heater vacuum sweeper Air condition Electric Fan Electric gas bottle Kerosene heater electric Iron colored Television Black white Television Video Cassette Normal stereo Radio Dish Cable Personal Computer Normal Camera Video Camera Water Filter Page 81 7 4 T a b l e A . 2 . 2 7 : S h a r e o f V a r i o u s E x p e n d i t u r e I t e m s t o T o t a l E x p e n d i t u r e b y P o v e r t y S t a t u s 2 0 0 5 u l t r a p o o r P o o r V u l n e r a b l e O t h e r s T o t a l u l t r a p o o r P o o r V u l n e r a b l e O t h e r s T o t a l u l t r a p o o r P o o r V u l n e r a b l e O t h e r s T o t a l E x p e n d i t u r e o n f o o d 5 5 . 2 2 5 3 . 9 2 5 2 . 5 6 4 3 . 3 5 4 4 . 5 0 5 7 . 8 2 5 6 . 4 8 5 4 . 8 7 5 1 . 4 3 5 2 . 9 7 5 7 . 3 1 5 5 . 9 0 5 4 . 0 9 4 6 . 2 1 4 8 . 1 8 E x p e n d i t u r e o n c i g a r e t t e s 3 . 6 6 3 . 4 4 3 . 9 5 2 . 6 7 2 . 8 0 2 . 8 1 3 . 2 0 3 . 4 0 3 . 0 9 3 . 1 7 2 . 9 8 3 . 2 6 3 . 5 9 2 . 8 1 2 . 9 6 E x p e n d i t u r e o n h o u s i n g 1 6 . 2 2 1 5 . 4 7 1 4 . 3 3 1 5 . 1 0 1 5 . 0 5 1 6 . 9 5 1 6 . 5 9 1 7 . 0 5 1 7 . 9 5 1 7 . 5 4 1 6 . 8 0 1 6 . 3 3 1 6 . 1 3 1 6 . 1 1 1 6 . 1 3 E x p e n d i t u r e o n c l o t h i n g 7 . 8 1 8 . 5 1 8 . 4 6 8 . 1 6 8 . 1 9 7 . 7 6 8 . 2 5 8 . 3 7 8 . 0 6 8 . 1 6 7 . 7 7 8 . 3 1 8 . 4 0 8 . 1 2 8 . 1 8 E x p e n d i t u r e o n f u r n i t u r e 3 . 7 2 3 . 5 9 3 . 5 4 4 . 7 6 4 . 6 2 3 . 5 6 3 . 5 3 3 . 6 6 4 . 2 7 4 . 0 2 3 . 5 9 3 . 5 4 3 . 6 2 4 . 5 9 4 . 3 6 E x p e n d i t u r e o n h e a l t h 3 . 0 2 3 . 0 7 3 . 4 3 4 . 7 1 4 . 5 4 2 . 6 2 2 . 8 4 2 . 9 1 3 . 8 2 3 . 4 7 2 . 7 0 2 . 8 9 3 . 0 9 4 . 3 9 4 . 0 8 E x p e n d i t u r e o n t r a n s p o r t a t i o n 2 . 5 6 2 . 9 7 3 . 3 2 5 . 4 0 5 . 1 4 2 . 5 1 2 . 5 2 2 . 7 0 3 . 1 1 2 . 9 3 2 . 5 2 2 . 6 2 2 . 9 1 4 . 5 9 4 . 1 8 E x p e n d i t u r e o n c o m m u n i c a t i o n 0 . 6 7 1 . 2 8 1 . 8 3 3 . 4 6 3 . 2 5 0 . 3 8 0 . 6 1 0 . 9 0 1 . 5 2 1 . 2 4 0 . 4 4 0 . 7 6 1 . 2 2 2 . 7 8 2 . 3 8 E x p e n d i t u r e o n e d u c a t i o n 2 . 1 0 2 . 3 0 2 . 7 2 4 . 4 5 4 . 2 3 1 . 6 3 1 . 8 6 1 . 9 2 1 . 9 4 1 . 9 2 1 . 7 2 1 . 9 6 2 . 1 9 3 . 5 6 3 . 2 3 E x p e n d i t u r e o n r e c r e a t i o n 1 . 2 5 1 . 4 2 1 . 7 0 3 . 7 4 3 . 4 9 0 . 9 2 1 . 1 0 1 . 2 4 1 . 7 2 1 . 5 2 0 . 9 8 1 . 1 7 1 . 4 0 3 . 0 3 2 . 6 3 E x p e n d i t u r e o n h o t e l s 0 . 1 1 0 . 1 2 0 . 1 3 0 . 1 6 0 . 1 5 0 . 0 8 0 . 1 0 0 . 1 2 0 . 1 3 0 . 1 2 0 . 0 8 0 . 1 1 0 . 1 2 0 . 1 5 0 . 1 4 O t h e r e x p e n d i t u r e 3 . 6 7 3 . 9 2 4 . 0 0 4 . 0 4 4 . 0 3 2 . 9 6 2 . 9 1 2 . 8 6 2 . 9 7 2 . 9 4 3 . 1 0 3 . 1 4 3 . 2 5 3 . 6 6 3 . 5 6 T o t a l a c t u a l c o n s u m p t i o n 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0 U r b a n R u r a l T o t a l Page 82 75 Table A.2.28: Fertility Rate and Under Five Mortality Rate by Poverty Status, 2004-05. Table A.2.29: Unemployment Rate of Youth (15-24 years) by Educational Status and Poverty, 2005. Ur b a n Ru r a l Al l Egyp t Non Poor Poor Non Poor Poor Non Poor Poor Illiterate Can read and write Basic Education secondary degree or equivalent Higher than Secondary degree but below university degree university degree and higher All Fer t ility Rat e Un d e r Fi v e Mo r t al i ty Rat e Ur b a n N on P oor Poor Total Ru r a l Non Poor Poor Total Al l Egyp t Non Poor Poor Total Page 83 76 Table A.2.30: Net Enrolment Rate by School Type and Poverty Status for Different Levels of Education, 2004-05. Primary Schools Preparatory Schools Net En r o l lm e n t Rat e Gi r l s Net En r o l lm e n t Rat e Net En r o l lm e n t in Pu bl i c Sch ool s Net En r o l lm e n t Rat e Gi r l s Net En r o l lm e n t Rat e Net En r o l lm e n t in Pu bl i c Sch ool s Urban Areas Non Poor Poor Total Rural Areas Non Poor Poor Total All Egypt Non Poor Poor Total Secondary Schools Universities Urban Areas Non Poor Poor Total Rural Areas Non Poor Poor Total All Egypt Non Poor Poor Total Page 84 77 Table A.2.31: Regression of Log Welfare Measure (Consumption/Poverty Line) on Characteristics of Household and Household Head for 2004-05 and 1999-00. BStd Error B Std Error Percentage of employed persons in government Percentage of illiterate persons Percentage of employers in agriculture Have taped water inside dwelling Durable Goods index Have private kitchen Have a private car Durable Goods index Constant Crowdedness index Have private bathroom Have access to public water network Percentage of employed persons in formal private sector Percentage of university degree holders Mean years of schooling for persons years and above Gender Percentage of unemployed persons Percentage of employed persons Log household size Log household size squared Share of children to Share of elderly persons Have electricity Connected to sewerage system network Have proper means of garbage collection Wall of house made of bricks Head works a permenant job Head works in government sector Head works in private sector Head is a wage worker in agriculture Head can read and write Head has basic education degree Head has secondary education degree Head has a diploma Head has university degree Age of the Head Age of the Head squared Metropolitan Lower Urban Lower Rural Upper Urban Upper Rural Border Urban R Square Number of observations Page 85 78 Table A.2.32: Impact of Changes in Household Characteristics and Characteristics of the Household Head on Poverty. (Percent Change). Metropolitan Lower Urban Lower Rural Upper Urban Upper Rural New born child Head work in agriculture House connected to sewerage system head became unemployed increase number of employed persons by one Head has a secondary degree Head has a university degree Female headed households New born child Head work in agriculture House connected to swewrage system head became unemployed increase number of employed persons by one Head has a secondary degree Head has a university degree Female headed households Page 86 79 Table A.3.1: Exchange Rates and Consumer Prices, 2000-2005 (Annual Change, December over December) 2000 2001 2002 2003 2004 2005 Cumulative Trade Weighted Exchange Rate -3.0% -7.4% 0.1% 41.3% 4.4% -9.2% 26.2% Nominal Exchange Rate 8.1% 16.6% 3.3% 31.2% 1.0% -7.9% 52.2% Consumer Price Index 2.2% 2.4% 2.9% 6.2% 10.8% 3.1% 27.6% Table A.3.2 Disaggregated Price Change (Cumulative Growth Rate July 2000-June 2005) Log change in price index, 2005:6 over 2000:7 Lower Egypt Upper Egypt Lower Egypt Upper Egypt Cairo Alex Canal Border Urban Urban Rural Rural All Items 0.26 0.29 0.30 0.28 0.23 0.29 0.28 0.28 Food Beverage & Tobacco 0.38 0.39 0.41 0.37 0.41 0.39 0.36 0.34 Bread & Cereals 0.21 0.21 0.21 0.26 0.30 0.22 0.35 0.44 Meat & Pouitry 0.43 0.42 0.43 0.40 0.44 0.43 0.34 0.30 Fish 0.50 0.47 0.53 0.54 0.45 0.47 0.42 0.42 Milk & Cheese 0.38 0.41 0.40 0.41 0.45 0.44 0.49 0.46 Oil & Fats 0.43 0.40 0.42 0.40 0.45 0.42 0.39 0.36 Fruits 0.61 0.59 0.62 0.47 0.59 0.54 0.36 0.36 Vegetables 0.25 0.40 0.59 0.31 0.47 0.43 0.56 0.28 Pulses 0.34 0.33 0.42 0.32 0.35 0.37 0.31 0.34 Sugar & Sweets 0.38 0.37 0.17 0.44 0.40 0.41 0.32 0.34 Other Food Stuff 0.31 0.28 0.28 0.29 0.27 0.29 0.21 0.21 Beverages 0.23 0.24 0.35 0.23 0.25 0.26 0.23 0.23 Tobacco 0.31 0.31 0.42 0.30 0.31 0.30 0.28 0.28 Clothing & Footwear 0.22 0.24 0.23 0.21 0.27 0.27 0.24 0.27 Clothing 0.22 0.24 0.22 0.18 0.28 0.27 0.22 0.25 Fabrics 0.33 0.28 0.28 0.33 0.32 0.36 0.41 0.45 Footwear 0.21 0.25 0.29 0.31 0.25 0.26 0.21 0.22 Clothing manufacture 0.06 0.21 0.23 0.18 0.18 0.16 0.08 0.13 Rent, Power & Fuel 0.10 0.13 0.11 0.10 0.10 0.08 0.13 0.13 Rent & Water 0.11 0.15 0.13 0.11 0.11 0.12 0.13 0.14 Energy & Fuel 0.06 0.09 0.08 0.10 0.10 0.04 0.16 0.11 Furnture & Equipmet 0.26 0.25 0.28 0.21 0.22 0.25 0.21 0.21 Furnture 0.21 0.19 0.19 0.17 0.18 0.18 0.18 0.17 Maintenance Products 0.28 0.29 0.28 0.25 0.25 0.26 0.23 0.24 Domestic Services 0.29 0.26 0.73 0.22 0.23 0.26 0.19 0.19 Medical Care 0.12 0.19 0.21 0.22 0.18 0.11 0.18 0.16 Medical Products 0.16 0.16 0.16 0.17 0.16 -0.01 0.16 0.16 Physician & Hospitals 0.08 0.22 0.26 0.27 0.19 0.20 0.20 0.16 Transport & Communication 0.30 0.29 0.34 0.33 0.37 0.48 0.30 0.29 Private Transportation 0.21 0.21 0.25 0.26 0.24 0.13 0.25 0.24 Purchased Transportation 0.17 0.14 0.23 0.24 0.24 0.41 0.24 0.24 Communication 0.55 0.60 0.63 0.62 0.67 0.43 0.64 0.62 Recreation & Education 0.14 0.17 0.14 0.19 0.15 0.10 0.11 0.16 Equipments 0.13 0.12 0.10 0.11 0.12 0.12 0.11 0.12 Entertainment & Cult. Serv 0.14 0.20 0.15 0.23 0.16 0.16 0.12 0.21 Education 0.15 0.06 0.15 0.19 0.14 0.17 0.10 0.10 Miscellaneous 0.19 0.18 0.20 0.21 0.21 0.21 0.17 0.17 Personal Care 0.15 0.14 0.18 0.18 0.19 0.18 0.14 0.14 Restaurants Hotels 0.26 0.26 0.28 0.27 0.26 0.27 0.25 0.25 Mean 0.25 0.26 0.30 0.27 0.28 0.27 0.26 0.25 SD 0.13 0.12 0.16 0.12 0.13 0.13 0.13 0.11 Page 87 80 Table A.4.1: Estimated Per-Capita Region-Specific Poverty Lines (L.E. Per Year) for 1999/2000 and 2004/2005 Lower poverty line (LE. Per capita per year) Region 1999/2000 2004/2005 Governorates Metropolitan 1,109 1,453 Cairo, Alexandria, Port Said, Suez Lower Egypt Urban 1,015 1,430 Lower Egypt Rural 978 1,429 Damiette, Dakhalia, Sharkia, Kalyoubia, Kafr El-Shaikh, Gharbia, Menoufia, Behera, Ismaila Upper Egypt Urban 1,031 1,416 Upper Egypt Rural 964 1,408 Giza, Beni-Suef, Fayoum, Menia, Assyout, Suhag, Quena, Aswan Page 88 81 Table A.4.2: Employment Structure and Growth Rate by Type of Employment, Sex and Urban/Rural Location, 1998-2006, Age 15-64 (in thousands) Av Ann Av Ann Gr percent Gr percent Ma l e Government Public Enterprises Formal Private Regular Wage Informal Private Regular Wage Irregular Wage To t al Wag e Wor k H H E nterprise W orker S elf E mployed T ot a l N on W ag e W or k T ot a l E mp l oy m e n t A ctive U nemployed Student out of Labor Force Non Student out of Labor Force Out of Man Power To t al Not Wor ki n g All Males Fem a l e Government Public Enterprises Formal Private Regular Wage Informal Private Regular Wage Irregular Wage To t al Wag e Wor k HH Enterprise Worker Self Employed To t al Non Wag e Wo r k To t al Em p l oy m e n t Active Unemployed Student out of Labor Force Non Student out of Labor Force Out of Man Power To t al Not Em p l oy ed All Females Bot h Government Public Enterprises Formal Private Regular Wage Informal Private Regular Wage Irregular Wage To t al Wag e Wor k HH Enterprise Worker Self Employed To t al Non Wag e Wo r k To t al Em p l oy m e n t Active Unemployed Student out of Labor Force Non Student out of Labor Force Out of Man Power To t al Not Em p l oy ed All Egypt Ty p e of Em p l oym e n t Av Ann Gr percent 1998-2006 A ges i n t hou s a nd s Ur b a n Ru r a l Al l Eg yp t Employment Structure and Growth Rate by Type of Employment, Sex and Urban/Rural Location, Page 89 82 Table A.4.3: Employment Structure and Growth Rate by Economic Activity, Sex and Urban/Rural Location 1998-2006 S ect o r o f A ct i vi t y Ma l e Agriculture Fishing Mining Manuf Utilities Construction Trade Hotels Restaurants Transp Storage Comm Financial Business Services Public Services Other T ot a l E mp l oyed T ot a l N ot E mp l oyed All Males F em a l e A griculture F ishing M ining M anuf U tilities Construction Trade Hotels Restaurants Transp Storage Comm Financial Business Services Public Services Other To t al Em p l oyed To t al Not Em p l oyed All Females Bot h Agriculture Fishing Mining Manuf Utilities Construction Trade Hotels Restaurants Transp Storage Comm Financial Business Services Public Services Other To t al Em p l oyed To t al Not Em p l oyed All Egypt A v A nn Gr percent Av Ann Gr percent Av Ann Gr percent Employment Structure and Growth Rate by Economic Activity, Sex and Urban/Rural Location, 1998-2006 A ges 15-64, in thousands Ur b a n Ru r a l Al l Egyp t Page 90 83 Table A.4.4: Cross-Sectional and Longitudinal Method of Calculating the Growth in Agriculture Wage and Agriculture Non-Wage Work by Sex and Urban/Rural Location 1998-2006 (in thousands) Sect or of Ac t ivi t y Av Ann Gr percent A v A nn Gr percent A v A nn Gr percent Ma l e Agr Wage Work Agr Non Wage Work Fem a l e Agr Wage Work Agr Non Wage Work Bot h Agr Wage Work Agr Non Wage Work Sector of Activity 1998 2006 1998 2006 1998 2006 Male Agr. Wage Work 183 119 -5.7 1,923 993 -8.8 2,106 1,112 -8.5 Agr. Non-Wage Work 108 172 6.2 1,452 2,382 6.6 1,561 2,554 6.6 Female Agr. Wage Work 53 12 -19.4 515 112 -20.4 568 124 -20.3 Agr. Non-Wage Work 81 122 5.4 1,072 1,475 4.3 1,153 1,597 4.3 Both Agr. Wage Work 236 131 -7.8 2,438 1,105 -10.5 2,674 1,237 -10.3 Agr. Non-Wage Work 189 294 5.9 2,524 3,857 5.7 2,714 4,151 5.7 A Cross sectional Method Urban Rural All Egypt Start Working before Av. Ann Gr, percent Start Working before Av. Ann Gr, percent Start Working before Av. Ann Gr, percent B. Longitudinal Method Urban rural All Egypt Page 91 84 Table A.4.5: Distribution of Real Monthly Earnings for Wage and Salary Workers by Background Characteristics (2006=100), 1988-2006 (using the FPI) N OTE : As mentioned in the main report, for the sake of comparability, all 1988 and 1998 monetary wages are inflated to 2006 Egyptian pounds using both CPI and FPI. The earning tables and figures reported in the previous sections of the paper uses the CPI inflation factor. The equivalent tables and figures using the FPI are summarized in this appendix. The FPI Inflation factor applied in this analysis is 1.65 from 1998 to 2006, and 4.83 from 1988 to 2006 background characteristics 1988 1998 2006 1988-1998 1998-2006 Total 483 359 415 -3 2 Gender Male 523 371 430 -3.4 2 Female 362 329 377 -0.9 1.8 Age group 15-24 290 247 295 -1.6 2.4 25-34 406 329 390 -2.1 2.3 35-44 583 397 440 -3.8 1.4 45-54 724 494 547 -3.8 1.4 55-64 660 505 630 -2.7 3.1 Region Urban Governorates 591 494 520 -1.8 0.7 Urban Lower Egypt 491 407 433 -1.9 0.9 Rural Lower Egypt 382 313 375 -2 2.5 Urban Upper Egypt 471 428 493 -0.9 1.9 Rural Upper Egypt 390 255 349 -4.2 4.3 Education Level Illiterate 390 265 333 -3.9 3.2 Literate without diploma 487 329 375 -3.9 1.8 Elementary school 503 346 361 -3.7 0.6 Middle school 487 357 435 -3.1 2.7 General high school 752 461 480 -4.9 0.5 Vocational high school 402 329 390 -2 2.3 Post-secondary institute 487 395 460 -2.1 2.1 University or higher 696 544 567 -2.5 0.6 Working Hours Per Week Median hours>=35 507 395 433 -2.5 1.3 Median hours < 35 402 231 300 -5.6 3.6 Median Real Monthly Earnings Level (in 2006 L.E.) Av. Ann Gr. percent Page 92 85 Table A.4.6: Distribution of Real Monthly Wage for Wage and Salary Workers by Institutional Sector and Economic Activity (2006=100), 1998-2006 (Using FPI) Sector Males Female Total Males Female Total Males Female Total 1998 2006 Sector of Activity Agriculture & Fishing 250 102 236 300 140 286 2.5 4.3 2.6 2.4 2.1 Mining, Manufacturing & Utilities 452 321 435 467 250 450 0.4 -3.5 0.5 1.4 1.9 Construction 329 247 329 390 798 390 2.3 16.2 2.3 1.3 0.5 Trade, Hotels & Restaurants 412 231 404 417 250 400 0.2 1.1 -0.1 1.8 1.7 Transportation, Storage & Communication 494 494 494 542 638 550 1.3 3.5 1.5 1 0.8 Financial & Business Services 646 577 626 650 500 600 0.1 -2 -0.6 1.1 1.3 Public Services 366 337 353 440 408 427 2.5 2.6 2.6 1.1 1.1 Other 300 329 300 300 320 300 0 -0.4 0 0.9 0.9 Institutional sector Government 367 346 362 438 429 435 2.4 3 2.5 1.1 1 Public Enterprises 533 539 533 583 516 581 1.3 -0.6 1.2 1 1.1 Formal Private Regular Wage 537 412 527 596 370 550 1.4 -1.5 0.6 1.3 1.6 Informal Private Regular Wage 428 214 412 390 200 390 -1.3 -0.9 -0.8 2 2 Irregular Wage 224 99 217 280 135 261 3.1 4.3 2.5 2.3 2.1 1998 2006 1998-2006 Males/ Females Median Real Monthly Wage Level (in 2006 L.E.) % change Gender Wage Ratio Page 93 86 Table A.4.7: Share of Low Monthly Wage Earners, Wage and Salaried Workers 1998-2006 % change 1998 2006 1998-2006 Total 53 45.5 -2.1 Gender Male 51.3 43.9 -2.1 Female 60 51.9 -2 Age group 15-24 74.9 69.9 -0.9 25-34 59.1 48.7 -2.7 35-44 48.6 40.4 -2.5 45-54 33.2 26.1 -3.3 55-64 37.6 25.3 -5.5 Region Urban Governorates 34.3 29.2 -2.2 Urban Lower Egypt 45.4 40.5 -1.6 Rural Lower Egypt 66.6 60.6 -1.3 Urban Upper Egypt 39.8 33.5 -2.4 Rural Upper Egypt 70.6 54.9 -3.5 Education Level Illiterate 73.4 41 -8 Literate without diploma 56.7 49 -2 Elementary school 55.8 44.3 -3.2 Middle school 50.8 58.3 1.9 General high school 26.5 60.3 11.3 Vocational high school 57.8 48.9 -2.3 Post-secondary institute 49 64.4 3.8 University or higher 27.6 72.8 13.4 Working Hours Per Week Median hours>=35 49.7 43.2 -1.9 Median hours < 35 76.8 65.8 -2.1 Sector of Activity Agriculture & Fishing 76 73.5 -0.4 Mining, Manufacturing & Utilities 40.4 41.6 0.4 Construction 59.3 48.9 -2.7 Trade, Hotels & Restaurants 49.7 46.3 -1 Transportation, Storage & Communication 36.8 27.1 -4.2 Financial & Business Services 23.7 29.8 3.2 Public Services 56 43.2 -3.6 Other 61.7 63.6 0.4 Institutional sector Government 54.5 41.8 -3.6 Public Enterprises 28.4 25.1 -1.7 Formal Private Regular Wage 28.5 28.8 0.2 Informal Private Regular Wage 50.6 53.3 0.7 Irregular Wage 82.5 53.3 -6 Total number of wage earners (000) 10,812 13,756 3.3 Total WAP (000) 36,800 44,900 2.7 Share of wage earners with low earnings* Level (2006=100) percent Page 94 87 Table A.4.8: Transition Across Low/High Earnings by Sex, 1998, 2006 from Wage Employment in 1998 to Wage Employment in 2006 (Using FPI) Table A.4.9: Transition Across Low/High Earnings by Institutional Sector, 1998, 2006 (Using FPI) Stay Hi Stay Low Hi to Low Low to Hi Initally Hi Initially low Government Employment 31.8 22.6 2.5 26.6 10.0 6.6 100 44.3 55.7 46.3 Public Enterprise 39.9 9.8 7.9 12.4 21.7 8.4 100 69.5 30.5 9.7 Formal Private Regular Wage Work 50.0 8.5 4.6 13.5 13.6 9.7 100 68.2 31.7 9.7 Inform Private Regular Wage Work 16.4 16.7 10.2 15.5 17.3 23.9 100 43.9 56.2 16.9 Irregular Wage Work 3.5 33.5 4.7 16.9 7.9 33.6 100 16.1 83.9 17.4 Total Wage Employed in 1998 26.7 20.8 4.9 20.3 12.3 14.6 100 43.9 55.7 100.0 Distribu- tion in 1998 Stayers Total Table A5.9: Transition Across Low/High Earnings by Institutional Sector 1998, 2006 Sector Movers Hi to Other Low to Other Transition from 1998 for Wage Employed Stay Hi S tay Low H i to Low Low to Hi missing Male 26.3 21.3 5.6 20.1 13.0 13.4 0.4 100.0 54.7 Female 28.3 19.1 2.2 21.2 9.8 19.0 0.3 100.0 59.4 T otal Wage Employed in 1998 2 6.7 20.8 4.9 20.3 12.3 14.6 0.4 100.0 55.7 S tay Hi Stay L ow Hi to L ow Low to Hi missing M ale 20.5 16.6 4.4 15.7 17.7 24.8 0.3 100.0 45.8 Female 22.5 15.2 1.7 16.9 11.6 31.9 0.1 100.0 48.9 T otal Wage Employed in 2006 20.9 16.3 3.8 15.9 16.4 26.3 0.3 100.0 46.5 Table A5.8: Transition Across Low/High Earnings by Sex 1998, 2006, to Wage Employment in 2006 E conomic Activity in 2006 T ransitions to 2006 Stayers Movers Other to H i Other to L ow Total P ercent L ow Pay in 1998 Percent Low Pay i n 2006 T able A5.7: Transition Across Low/High Earnings by Sex 1998, 2006, from Wage Employment in 1998 Economic Activity in 1998 Transitions from 1998 Stayers Movers H i to Other L ow to Other Total Page 95 88 A NNEX F IGURES Figure A.1.1: Predicted Poverty Rates at Village Level and Their Confidence Intervals; in Rural Areas, 1996 Figure A.1.2: Predicted Poverty Rates at the Sub-District Level and Their Confidence Intervals; in Urban Areas, 1996 -20 0 20 40 60 80 100 120 p0 p0+1.96se p0-1.96se -40.0 -20.0 0.0 20.0 40.0 60.0 80.0 100.0 1 10 19 28 37 46 55 64 73 82 91 100 109 118 127 136 145 154 163 172 181 p0 p0+1.96se p0-1.96se Page 96 89 0 . 0 5 . 1 . 1 5 c v d i r e c t 7 8 9 10 11 lnexptotal Figure A.3.1: Distribution of Estimated Long-Run Exchange Rate Pass-Through to Consumer Prices Aggregate Items - . 5 0 . 5 1 A l l C l o t h i n g & F o o t w e a r E d u c a t i o n E n t e r t a i n m e n t F o o d B e v e r a g e s T o b a c c o F u r n i t u r e & E q u i p m e n t M e d i c a l C a r e M i s c e l l a n e o u s R e n t & P o w e r & F u e l R e s t a u r a n t s & H o t e l s T o b a c c o T r a n s p o r t & C o m m u n i c a t i o n Disaggregated Food Items 0 . 2 . 4 . 6 . 8 1 B e v e r a g e s B r e a d & C e r e a l s F i s h F o o d B e v e r a g e s T o b a c c o F r u i t s M e a t & P o u l t r y M i l k & C h e e s e O i l & F a t s O t h e r F o o d S u g a r & S w e e t s V e g e t a b l e s Figure A.3.2: Direct Effects of Price Changes on Welfare (Compensating Variation Calculated as Percent Change in Total Expenditure Required to Purchase Initial Consumption Basket) Page 97 90 Figure A.4.1: Distribution of Real Monthly Earnings in Relation to a Low Earnings Threshold by Sex, 1998-2006 (Using CPI) 0 . 0 0 0 5 . 0 0 1 . 0 0 1 5 . 0 0 2 P e r c e n t 0 1000 2000 3000 4000 R eal wage 1998 2006 P L ( Feb. 2006) =374. 3 9 L E R eal Wa g es Cal c ul a t ed Us i ng FPI D ist r i but i on o f Real Mo n t hl y Wa g e, ( 2006=100) , Ag e s 1 5- 6 4, 1998, 2006 0 . 0 0 0 5 . 0 0 1 . 0 0 1 5 . 0 0 2 P e r c e n t 0 1000 2000 3000 4000 Real wage 1998 2006 PL ( Feb. 2006) =374. 3 9 L E Real Wa g es Cal c ul a t ed Us i ng FPI Ma l e Di s t ri but i on o f Real Mo n t hl y Wa g e (2006=100) , Ag e s 1 5- 6 4, 1998, 2006 0 . 0 0 0 5 . 0 0 1 . 0 0 1 5 . 0 0 2 . 0 0 2 5 P e r c e n t 0 500 1000 1500 2000 Real wage 1998 2006 PL ( Feb. 2006) =374. 3 9 L E Real Wa g es Cal c ul a t ed Us i ng FPI Fema l e Di s t ri but i on o f Real Mo n t hl y Wa g e (2006=100) , Ag e s 1 5- 6 4, 1998, 2006 0 . 2 . 4 . 6 . 8 1 C . D . F . 0 1000 2000 3000 4000 R eal Wa g e 1998 2006 P L ( Feb. 2006) = 368. 2 7 L E R eal Wa g es Cal c ul a t ed Us i ng FPI C umu l at i ve Di s t ri but i on o f Real Mo n t hl y Wa g e ( 2006=100) , Ages 15- 6 4, 1998, 2006 0 . 2 . 4 . 6 . 8 1 C . D . F . 0 1000 2000 3000 4000 Real Wa g e 1998 2006 PL ( Feb. 2006) = 368. 2 7 L E Real Wa g es Cal c ul a t ed Us i ng FPI Ma l e Cu m u l at i ve Di s t ri but i on o f Real Mo n t hl y Wa g e (2006=100) , Ages 15- 6 4, 1998, 2006 0 . 2 . 4 . 6 . 8 1 C . D . F . 0 500 1000 1500 2000 Real Wa g e 1998 2006 PL ( Feb. 2006) = 368. 2 7 L E Real Wa g es Cal c ul a t ed Us i ng FPI Fema l e Cumu l at i ve Di s t ri but i on o f Real Mo n t hl y Wa g e (2006=100) , Ages 15- 6 4, 1998, 2006 Page 98 91 Figure A.4.2: Distribution of Real Monthly Earnings in Relation to a Low Earnings Threshold by Institutional Sector of Employment, 1998-2006 (Using the CPI). 0 . 2 . 4 . 6 . 8 1 0 1000 2000 3000 GovEmploy 0 . 2 . 4 . 6 . 8 1 0 1000 2000 3000 PubEnt 0 . 2 . 4 . 6 . 8 1 0 1000 2000 FormalPriv_RegWage 0 . 2 . 4 . 6 . 8 1 0 1000 2000 InformPriv_RegWage 0 . 2 . 4 . 6 . 8 1 0 1000 2000 IrrWage 1998 2006 C . D . F . Real Wage PL (Feb. 2006)=368.27 LE Real Wages Calculated Using CPI Cumulative Distribution of Real Monthly Wage (2006=100), by Institutional Sector, Ages 15-64, 1998, 2006 0 . 0 0 1 . 0 0 2 . 0 0 3 0 500 1000 1500 2000 GovEmploy 0 . 0 0 1 . 0 0 2 0 500 1000 1500 2000 PubEnt 0 . 0 0 1 . 0 0 2 0 500 1000 1500 2000 FormalPriv_RegWage 0 . 0 0 1 . 0 0 2 0 500 1000 1500 2000 InformPriv_RegWage 0 . 0 0 2 . 0 0 4 0 500 1000 1500 2000 IrrWage 1998 2006 P e r c e n t Real wage PL (Feb. 2006)=368.27 LE Real Wages Calculated Using CPI Distribution of Real Monthly Wage (2006=100) by Institutional Sector, Ages 15-64, 1998, 2006 Page 99 92 FigureA.4.3: Distribution of Real Monthly Earnings in Relation to a Low Earnings Threshold, 1998-2006 (Using the FPI) 0 . 0 0 0 5 . 0 0 1 . 0 0 1 5 . 0 0 2 P e r c e n t 0 1000 2000 3000 4000 Real wage 1998 2006 PL (Feb. 2006)=374.39 LE Real Wages Calculated Using FPI Distribution of Real Monthly Wage, (2006=100), Ages 15-64, 1998, 2006 0 . 0 0 0 5 . 0 0 1 . 0 0 1 5 . 0 0 2 P e r c e n t 0 1000 2000 3000 4000 Real wage 1998 2006 PL (Feb. 2006)=374.39 LE Real Wages Calculated Using FPI Male Distribution of Real Monthly Wage (2006=100), Ages 15-64, 1998, 2006 0 . 0 0 0 5 . 0 0 1 . 0 0 1 5 . 0 0 2 . 0 0 2 5 P e r c e n t 0 500 1000 1500 2000 Real wage 1998 2006 PL (Feb. 2006)=374.39 LE Real Wages Calculated Using FPI Female Distribution of Real Monthly Wage (2006=100), Ages 15-64, 1998, 2006 0 . 2 . 4 . 6 . 8 1 C . D . F . 0 1000 2000 3000 4000 Real Wage 1998 2006 PL (Feb. 2006)=368.27 LE Real Wages Calculated Using FPI Cumulative Distribution of Real Monthly Wage (2006=100), Ages 15-64, 1998, 2006 0 . 2 . 4 . 6 . 8 1 C . D . F . 0 1000 2000 3000 4000 Real Wage 1998 2006 PL (Feb. 2006)=368.27 LE Real Wages Calculated Using FPI Male Cumulative Distribution of Real Monthly Wage (2006=100), Ages 15-64, 1998, 2006 0 . 2 . 4 . 6 . 8 1 C . D . F . 0 500 1000 1500 2000 Real Wage 1998 2006 PL (Feb. 2006)=368.27 LE Real Wages Calculated Using FPI Female Cumulative Distribution of Real Monthly Wage (2006=100), Ages 15-64, 1998, 2006 Page 100 93 Figure A.4.4: Distribution of Real Monthly Earnings in Relation to a Low Earnings Threshold by Institutional Sector of Employment, 1998-2006 (Using the FPI) 0 . 0 0 1 . 0 0 2 0 500 1000 1500 2000 GovEmploy 0 . 0 0 0 5 . 0 0 1 . 0 0 1 5 0 500 1000 1500 2000 PubEnt 0 . 0 0 0 5 . 0 0 1 . 0 0 1 5 0 500 1000 1500 2000 FormalPriv_RegWage 0 . 0 0 1 . 0 0 2 0 500 1000 1500 2000 InformPriv_RegWage 0 . 0 0 1 . 0 0 2 . 0 0 3 0 500 1000 1500 2000 IrrWage 1998 2006 P e r c e n t Real wage PL (Feb. 2006)=374.39 LE Real Wages Calculated Using FPI Distribution of Real Monthly Wage (2006=100) by Institutional Sector, Ages 15-64, 1998, 2006 0 . 2 . 4 . 6 . 8 1 0 1000 2000 3000 GovEmploy 0 . 2 . 4 . 6 . 8 1 0 1000 2000 3000 PubEnt 0 . 2 . 4 . 6 . 8 1 0 1000 2000 3000 FormalPriv_RegWage 0 . 2 . 4 . 6 . 8 1 0 1000 2000 InformPriv_RegWage 0 . 2 . 4 . 6 . 8 1 0 1000 2000 3000 IrrWage 1998 2006 C . D . F . Real Wage PL (Feb. 2006)=368.27 LE Real Wages Calculated Using FPI Cumulative Distribution of Real Monthly Wage (2006=100), by Institutional Sector, Ages 15-64, 1998, 2006