chapter2

Every poverty assessment faces a straightforward, but complicated question. That is, how to define and measure poverty. In its simplest terms, the question boils down to this: How poor is poor? There is some guiding consensus on the issue of what poverty means, which is reflected in the Human Development approach. Poverty is more than just lack of food as it involves human deprivation in the consumption of other commodities and services. But what are these services and in what quantity? Is the lack of means to buy an interesting book or visit the local theatre signs of poverty? Many would answer no, as one can physically survive without these things. However, for others who derive great pleasure from the consumption of these goods, the answer would be "Yes". Is the lack of means to provide a proper burial to a family member a sign of extreme poverty? Many would answer "Yes" to this question while others would argue that it is just another luxury that one can live without. Where should we draw the line?

The definition of "poverty" carries significant consequences for the poverty headcount in Georgia (the number of people below and above the poverty line) and therefore it bears some political sensitivity. The per capita spending of approximately 40% of the population falls within the narrow range of 55 GEL and 115 GEL per month, a range that coincides with the lowest and highest poverty lines currently applied in Georgia. Because the expenses of the population are bunched in a narrow interval, moving the poverty line up or down carries a disproportional effect on the poverty headcount.

Poverty lines are affected by non-trivial doses of subjectiveness. Any method applied in the construction of a poverty line reflects implicit or explicit differences in the understanding of "poverty" and therefore different methods are likely to result in different poverty thresholds. There is nothing like the "right", "scientific" or "objective" poverty line. Even the application of sophisticated econometric models to define poverty lines is not silent about the implicit beliefs and assumptions embedded in each of them. It would be disingenuous to declare otherwise.

Because poverty lines express particular beliefs of what poverty means, care should be exercised in reading too much from them. To claim high moral ground because one uses a poverty line that is higher than another is hardly a useful way of framing a discussion that, in the end, is about personal beliefs and expectations. On the other hand, to claim that a poverty line is "more realistic" than another because it is lower, identifies a smaller number of "poor", and therefore may allow better targeting of scarce government resources, could hardly be seen as a monumental thought. First, if the argument is duly followed all the way

down, one should determine a poverty threshold not by notions of deprivation or need but rather by the availability of government funds to combat poverty. Second, if the concern is the difficulty of constructing a poverty profile that serves to identify groups in direst need, one can still set a poverty line anchored on notions of need and deprivation and then undertake poverty assessments for groups that meet, let us say, 75% of the poverty line, or 50% of the poverty line, or 35% of the poverty line. To justify lower poverty lines on grounds of better targeting of scarce government resources reflects some confusion about notions of individual needs and deprivations with notions about the fiscal standing of the country in question.

As part of defining a poverty threshold, this chapter will present different alternative formulations for a poverty line. In doing this, the objective is not to determine which one reflects the "right" definition of poverty. Rather, the objective is to make transparent what different poverty lines imply in terms of options and choices available to the person that must make ends meet within them.

Because an element of arbitrariness is always present in the construction of poverty lines, different experts can come up with different poverty thresholds. This has been experienced in Georgia. So far, the definition of the poverty line has produced not just one line, but four. These are the following:

The Official Poverty Line is based on the cost of the Official Minimal Food Basket and determines the minimum subsistence level of a male of working age¹⁰. The cost of the food basket, which provides for 2,500 Kcal/day, represents 70% of the poverty line while the rest is assigned to non-food items. The "Revised Poverty Line" follows a similar approach but uses a food basket of less caloric content (2,200 Kcal/day) and shows a slight variation in the proportion of food and non-food items (65:35).

In turn, the "Recommended Poverty Line", which was developed by the World Bank in cooperation with the State Department of Statistics, the Ministry of Economy, and the Ministry of Social Protection, Labor and Employment, follows a similar approach than the official poverty line but it uses an even cheaper food basket of products. It is a sort of semi-official poverty line of great influence in policymaking and closely matches the value of the "Alternative Minimum" poverty line, which constitutes the main poverty threshold used in the "Poverty Reduction and Economic Growth Program".

Box 2.1: a personal perception of poverty

"People do everything to survive – they sell goods that they have in the household, but now almost all people have already sold everything. So you may call "poor" those who still have something to sell, and "poorest" those who already have nothing to sell" (Refugee. Tbilisi.)

All these poverty lines allocate resources for food and non-food consumption. As previously mentioned, the "food component" is simply a basket of food products that should provide for proper nourishment. The "non-food component" is money that may be used to meet needs such as health care, electricity, gas, heating, education, etc. Each poverty line provides allocations for food and non-food items, which together should ensure a minimum consumption level compatible with some shared notion of basic needs. The rest of this chapter will describe different potential formulations for a basket of food and non-food products that meets minimum basic needs, however these are defined.

The food component of a poverty line is affected by several factors. The most important is the composition of the food basket (or menu of food), which should provide sufficient calories, vitamins, proteins and micronutrient intake to keep a person properly nourished while attaining the objective at a minimum cost. Another factor is the time frame during which poor people are on the food basket. For instance, one can construct a food basket that while showing some deficiency in calories, proteins, or vitamins is still sufficient to ensure proper nourishment for, let us say, twelve months. The cost of the food basket should reflect the minimum cost of achieving a basic recommended calorie intake, which is often set at 2,500 Kcal/day. For many of our fellow citizens, achieving this minimum caloric threshold is not an easy goal, as Table 2.2 shows.

As Table 2.2 shows, the famous "Georgian Table", full of food and delicacies, is rapidly becoming a myth for an increasing number of fellow citizens. In Georgia, there are significant numbers of people who consume a diet that provides less than 1,800 Kcal a day. This is a poor diet that may increase susceptibility to illness and is only compatible with below normal (low) physical activity. The case of Tbilisi, which alone comprises 25% of the total population of Georgia, is particularly disturbing. A steady 39% to 40% of the population is improperly nourished.

Table 2.2 above also provides insight into the risks involved in taking poverty lines at face value. For example, according to the "Recommended Poverty Line", which provides for a food basket of 2,500 Kcal/day, the number of poor in Tbilisi is about 20%. Yet, according to Table 2.2, 40% of Tbilisi inhabitants are unable to meet 72% of the food basket. For Georgia as a whole, the "Recommended Poverty Line" indicates that 23% of people are poor but Table 2.2 tells that 45% of people cannot consume 88% of the food basket.

Two factors account for the discrepancies between poverty headcounts and percentage of people with insufficient calorie intake. One is that not all products in the food basket may be available in all regions throughout the year and therefore more costly substitutes may be needed. We will discuss this issue later in this chapter. The second, and perhaps most important, factor is that as the cost of the food basket is minimized, the order and content of consumption becomes quite sophisticated. In order to attain an adequate diet with approximately 38 GEL/month, which is the allowance for food in the "Recommended Poverty Line", then sophisticated knowledge about nutrition is needed because specific products should be consumed in carefully determined amounts. This is a level of knowledge that, unfortunately, most of us lack. As reported in roundtable discussions, participants declared unanimously that it would be "impossible" to make ends meet with 55 GEL/month even though it may be technically feasible. With 75 GEL/month, the answer was "hardly possible" while 110 GEL/month would provide a "more bearable existence".

For the purposes of this study, the NHDR has calculated the cost of a food basket that meets average recommended daily allowances for a number of different standards. The composition of the NHDR food basket, which is shown in Technical Annex 4, has been based on the basket used in the "Recommended Poverty Line". Basically, the NHDR food basket allows for a greater consumption of fruit, vegetables and meat products than in the "Recommended Poverty Line" and the "Alternative Minimum Poverty Line"¹¹. This increase in quantity and composition is not great but it would nevertheless make a difference for an increasing number of Georgians whom are falling below proper levels of food consumption as shown in Table 2.2. Finally, the protocol of consumption of the NHDR food basket has not been made too sophisticated so as to account for "mistakes" in consumption. The findings of Table 2.2 call for caution in devising sophisticated food baskets, simply because most people may not be able to follow them.

The calculation of the cost of the NHDR food basket presents an additional departure from the method being applied by the SDS and the World Bank. The "Recommended Poverty Line" and the "Alternative Minimum" use national average prices reported in the Consumer Price Index (CPI). The national CPI average is based on prices from the cities of Tbilisi, Kutaisi, Batumi, Gori and Telavi. There are some doubts as to whether these prices are representative of rural areas or other regions outside these cities. In contrast, the NHDR food basket uses prices directly reported by families from different regions of Georgia as reported by the SDS. The differences in prices can be noticeable and result in variations in the cost of the basket of as much as 20%. Prices reported by families are on average higher than prices reported in the CPI. There are two main advantages in using prices reported by families. First, these prices allow for the construction of different poverty lines by region, something that the CPI does not. Second, these prices appear to be the ones that families actually face. Using prices reported by families, the cost of the NHDR food basket across Georgian regions and between seasons is the following:

Finally, the cost of the food basket in Table 2.3 above has been adjusted to take into account family composition (equivalence scale adjustment) and family size (economies of scale adjustment). Briefly, an adjustment for equivalence scale obeys the fact it costs more to feed an adult than to feed a child. The adjustment for economies of scale is more controversial. Economies of scale in food consumption can only come true if large households are able to purchase food in bulk, thus obtaining price discounts. Food, we remember, can hardly be shared. In adjusting for economies of scale, the NHDR follows the approach set by the World Bank in its Poverty Report of 1999, which finds that an additional person in the household appears to reduce the cost of food by 8% with pensioners having the strongest effect on reducing the price of main food items (apparently because of low opportunity cost of searching for good deals)¹². The use of equivalence scales and adjustment for family size has the impact of reducing the poverty headcount because the cost of the food basket decreases for larger families with children.

The construction of the non-food component of the poverty line can demand an even higher degree of arbitrariness than that observed for the food component. At least in the case of food, there is an understanding of a minimum calorie and nutrient intake. Such an understanding is not available when attempting to define non-food requirements. Here is when we are back to the questions posed at the beginning of this chapter: How to determine which non-food commodities (and their respective quantities) are essential.

For example, is electricity essential? While for many Georgians this question is almost irrelevant (20% of the population reported to have had zero electricity supply in winter 2001), there may be some argument that a minimum amount of electricity is necessary. For instance, neither the NHDR nor the "Recommended Poverty Line" food basket remains much valid unless there is the possibility of storing food for some time. A good example is milk consumption. The 1.4 liters a month provided by the "Recommended Poverty Line" are supposed to be consumed in quantities of 46 grams a day, or half a cup of milk a day. Milk, however, has to be purchased in units of one liter and it is difficult to store in the absence of a refrigerator due to spoilage. Either the person can make use of a refrigerator or should consume her milk quite rapidly. The latter is not a good consumption pattern for children. This same constraint applies to the 2.13 liters a month allowed in the food basket of the NHDR.

A minimum consumption of gas (or other fuel) for cooking and heating also appears to be essential. There is little argument about the need for fuel to cook food. But what about keeping an apartment reasonably warm? In winter, it can have a great effect on overall health status¹³. In particular, the amount of calories in both the "Recommended Poverty Line" and NHDR line are not intended to counterbalance the effects of permanently cold household environments. The propensity to get sick because of chronic exposure to cold weather is not an unusual finding.

Education is another item that qualifies as important. In most societies education will open the door to higher long-term income. Among all levels of education, primary and secondary are the most crucial as they provide necessary, though not sufficient, endowments to avoid the perpetuation of poverty. But primary and secondary education are no longer free in Georgia. The last two grades of secondary school have a required fee of 10 GEL a month. In addition, we estimate that all school children spend approximately 6 GEL a month in educational expenses such as books, pencils, and informal contributions to the school¹⁴.

Health care is another item for which some level of spending might be needed. For a significant number of Georgians, health care remains beyond their means. The past years of reform and loans have not yet been successful at ensuring some basic provision of health care services. Informal fees in hospitals (under the table payments or "gifts" to doctors and nurses) are common currency and are a minimum of 2-3 GEL even for very basic services. Of course, this does not take into account the cost of medicine, for which a patient must pay herself. Neither the NHDR’s nor other food baskets are designed to cope with the effect of a debilitating illness. There are some risks in constructing a food basket that depends on the health status of the individual while divorcing oneself from the question of whether the individual has the means to pay for even a preventive visit to the doctor.

Surely there may be other items that the reader of this NHDR sees merit in including in a non-food basket. But there is also a need to keep the list of items from growing out of control. There are two potential avenues to deal with this issue. One is to approach the problem head-on and define what comes in and out of the non-food basket, a task heavily influenced by personal beliefs of what constitutes minimum non-food requirements. Though appealing because of its simplicity and straightforwardness, the direct definition of a non-food basket is not an avenue free of problems. First, the method is not widely used. In the republics of the former Soviet Union, only Uzbekistan has applied it consistently since 1992¹⁵. To some extent, there is value added in conducting poverty studies that make use of widely used methods because results can be compared across countries more easily. Second, there is a risk of being perceived as paternalistic. Who are we to decide what is necessary and what is not? To complicate matters, in the republics of the former Soviet Union, state paternalism is a touchy issue after decades of heavy state control on almost all aspect of social and economic life. Third, there is a degree of arbitrariness involved, as there is no objective standard similar to the minimum nutritional requirement to which our final choice of non-food items can be linked. Non-food baskets defined by the direct method are like opinions: everybody can have a different one.

At first glance, an attractive way out of this problem is to allow for a fixed amount and let the household decide what to do with it. This is what the Official, Revised and Recommended Poverty Lines do. This fixed amount varies between 35% and 30% of the cost of the poverty basket. For example, the Revised Poverty line is 75 GEL with a food to non-food share fixed at 65% and 35% respectively, which results in a food allowance of 49 GEL and a non-food allowance of 26 GEL.

Even though this method is attractive because it bypasses the definition of the non-food basket, it is not free of problems either. First, the amount allocated to non-food consumption is directly proportional to the "food base" on which it is being calculated. The lower the food basket, the lower the amount for non-food items. Second, the final allowance obtained for non-food items is hardly a meaningful figure in the absence of some analysis on what households are able to do with it^16,17.

Independent of the method chosen, the resulting poverty line cannot, and should not, escape scrutiny about the options and alternatives open to individuals that have to make ends meet within it. The direct definition of essential non-food item enters this debate straight on simply because it explicitly mentions what is included and what is not. The poverty line obtained by the use of fixed shares also enters the debate because the resulting non-food allowance is easily compared to the cost of visiting a doctor, sending a child to school and keeping an apartment relatively warm in winter (among other things). The comparison can be done without implying any order of importance to the expenses. Rather, the expenses simply reflect choices that individuals make based on the amount of money available for non-food expenditures.

This NHDR has estimated the cost of a minimum bundle of non-food items so as to allow an exploration of choices and alternatives open to individuals that fall under different poverty lines. Not all readers will agree with our choices, some because of too little, others because of too much. But there is a value added in this kind of debate. It makes transparent the discussion of options and alternatives open to individuals and facilitates understanding the kind of life lived by scores of fellow citizens. That is, how many people are giving up electricity consumption to have gas for cooking, or decreasing consumption in health care in order to pay for the education of a child, or diminishing ones own food consumption in order to pay for a warm apartment, and so on and so forth.

Let us see what one can do with the allocations in Table 2.4 and how they might vary according to family size. An allocation of 9.62 GEL for electricity allows a person to have 1 lamp of 100W turned on for five hours a day (1.49GEL/month), a refrigerator for five hours a day (3.72 GEL) and to produce a small tank of hot water once every four days (4.41 GEL). Of course, the individual is free to distribute these allocations in a different manner. For instance, she can increase the use of a refrigerator to seven hours a day and take a quick shower once a week. Apart from money, of course, this consumption of electricity demands at least five hours of electricity a day, something that last winter about 75% of Georgians could enjoy.

The 9.62 GEL expenditure on electricity (Table 2.4 above) has been computed based on the consumption of a single adult. In contrast to food, consumption of electricity allows for huge economies of scale so allowances do not grow proportionally to the number of household members. The NHDR has explored the increase in electricity spending with each additional household member for the bottom 30% of the consumption distribution. It was found that electricity spending stays almost unchanged with the first two members and on average increases by less than 5% with the third. A family of seven consumes on average44% more electricity than a family of two despite having more than three times the number of members. Families with more than seven members do not change this finding significantly.

In turn, an allocation of 12 GEL/month for gas and heat provides enough fuel to cook two meals a day and a heater turned on for four hours a day. Of course, the number of hours the heater can remain on will vary depending on the particular efficiency of the heater in question, but estimates suggest that a well-insulated room could stay relatively warm with four hours of heating a day¹⁸. As observed in the case of electricity, the increase in gas and heating expenses does not go up proportionally to the number of family members. There is no significant change in expenses with the addition of a second member. A third member, however, has a noticeable impact and can double expenses in gas/heating. The addition of a fourth member leaves expenses, on average, almost unchanged. A fifth member can increase expenses by 25% and the addition of two more members (for a family of 7) could add another 29%. After that, expenses show little change.

An allocation of 5 GEL a month for health care is insufficient to buy most remedies, but there are insurance health plans at a cost of 5 GEL a month. True, people may not have information about this type of insurance and the offer may not be available throughout all Georgia. For those unaware of the insurance option and whose health care should be restricted to 5 GEL a month, things can turn dire. Despite efforts aimed at establishing minimum standards of service within the public health care sector, a visit to a public hospital continues to demand informal payments with a minimum of 2-3 GEL in addition to any medicine that might be needed.

In terms of economies of scale, few, if any, are present in health care expenditures. A family member that falls sick can scarcely benefit from the expenses incurred on another family member in the same condition. Similarly, insurance is on a per-capita basis and every member that wishes to enroll would require 5 GEL. Because of this, if health care were chosen as an essential item in a non-food basket, then the allowance would need to be on a per-capita basis. Survey data do not provide much guidance about this issue because the bottom 30% of the population spends negligible amounts on health care. There are reasons to believe, however, that lack of household expenditure on health care reflects lack of money and neither countrywide good health conditions nor cost-free public health care services (see Chapter 3). For those that are unable to buy insurance, the 5 GEL allowance to each member could allow the pooling of resources at the intra-household level and relieve to some extent the family’s health care budget constraint. A cautious family of three whose members enjoy good health during the entire summer would be left with savings of approximately 45 GEL at the beginning of the winter period should any one of them require a visit to the doctor.

Table 2.4 also includes an allowance for transportation of 8.8 GEL, which would allow 22 trips at a cost of 20 tetri per ride. At first glance one could think of transportation as an unusual, or even inappropriate, item to include in a non-food basket. But some spending in transportation may be difficult to avoid. Take a Tbilisi resident for example. The breadwinner in a family has to go to work, which for most workers will demand the use of public services. Or take the example of a person who must search for the best food deals, an assumption applied when computing the costs of the NHDR and the "Recommended Poverty Line" food baskets. More often than not, these good deals are not available at ones doorsteps and require some traveling. There is some inconsistency in expecting at least one family member to work, or pensioner to find good deals for food, while divorcing oneself from the question of whether the individual can meet the cost involved in getting to work or finding the food deals. The survey data also shows that transportation is an expense even outside Tbilisi, Kutaisi, or Adjara. On average, Georgians spend on transportation half the amount spent on medicine and twice the amount spent on medical services (visits to hospitals).

If applied to families with more than one member, we estimate that no increase in transportation expenses will be needed up to the third member. For four members and above, an additional 8.8 GEL could be required. This reflects the observed composition of Georgian families, the likelihood of having a second working family member, and the need to search for good deals for food as the family gets bigger.

Finally, there might be a need to provide a minimum amount to cover the costs of education for children between the 1^st and 11^th grade. Education costs are both formal and informal. For children below 9^th grade, informal payments to the school and expenses for pencils, books and other items can easily amount to 6 GEL a month. For students above 9^th grade, an additional 10 GEL should be added in terms of school fees. As in the case of health care, there is little room for economies of scale in education expenses. Should education enter the list of essential non-food items, there would be a need for an allowance for every child in the household.

For a single adult male or female, the sum of items (electricity, gas, health care, transport) listed in Table 2.4 amounts to 35.4 GEL/month in winter and 26.92 GEL/month in summer. Because of the assumptions regarding economies of scale, as well as the selective allowance for education, the cost of meeting this bundle of non-food items will vary according to family size and composition¹⁹.

Let us turn attention to the options that may be open to an individual whose income gradually falls from 100 GEL/month to less than 50 GEL/month, which is the range of poverty lines presented in Table 2.1 at the beginning of this chapter.

A person living on approximately 100 GEL/month that follows the NHDR food basket spends approximately 65 GEL on food and 35 GEL in all other expenditures. This person has a difficult life. She needs to restrict light to five hours a day and turn off the heater after the fourth hour. There is no allowance for things like clothes, soap, a razor blade, toilet paper, a toothbrush, toothpaste, a book, or a present for a child, among other things. It is mandatory she take great care of her health as the allocation for health care is minimal and must be saved for the time it is really needed. If the work location is further than one bus ride, she has to either walk the remaining distance or cut down on other expenditures in order to make room for an extra bus ticket. On the other hand, she eats reasonably well and could remain on the diet for an extended time if need be. Expenditures on items other than those defined in Table 2.4 (Cost of monthly consumption of potential non-food basket) must come at the expense of reducing the quality of food consumption or reducing consumption of non-food items. Thus, any unexpected expense could result in a greater consumption of beans and bread, which still makes a stomach feel full, instead of fresh vegetables and fruits, which provide better vitamin and micronutrient intake.

As income goes down, she may gradually switch to the food basket of the "Recommended Poverty Line", which is cheaper than the NHDR’s, and costs approximately 42 GEL (using prices reported by families). This change in food baskets would free up to 24 GEL in winter

and 21 GEL in summer. Because of the higher poverty incidence in winter, such a strategy might be seen more often during cold months. However, the switch to a lower-cost food basket requires considerable knowledge about nutrition because as the diet is cut down, the greater the need to know very well which elements must be consumed and in which quantities.

By the time total consumption is down to 55 GEL a month, a person with professional knowledge about nutrition could have switched completely to the composition of the "Recommended Poverty Line" food basket. Good calorie and nutrient intake, however, still carry a penalty in terms of non-food consumption, particularly in winter. If she lives in Tbilisi, there are approximately 9.3 GEL left for whatever expenses remain. A refrigerator might be desirable for otherwise only non-perishable items, like bread, beans, macaroni, and potatoes can be consumed two-thirds of the month. This is not a good policy for children. The need to keep the apartment warm would alone consume more than the whole allowance for non-food items. A reasonably good diet and a refrigerator working five hours a day would demand a serious cut in heating expenditure, perhaps no expenditure at all for health care, nothing for transportation either, and negligible or zero expense on personal items (e.g. soap).

In summer, cutting down non-food expenditures is much less taxing in terms of family welfare because days are long and warm. An individual in Tbilisi with total consumption of 65 GEL/month that manages to follow the food basket of the "Recommended Poverty Line" would be left with approximately 20-23 GEL for other expenses. This is 74% of the total summer cost of the bundle of items in Table 2.4. Probably, among all options available, electricity may go unpaid as scores of Georgians do regularly. At the collective level, this is unfortunate since non-payment is one of the factors that has brought the energy system to its current sorry state (the other is gross mismanagement). By the time her income is down to 55 GEL/month, the situation becomes complicated even in summer. A person with professional knowledge of nutrition for whom a good diet is an overriding concern would be left with approximately 10 GEL for non-food expenses. Assuming that s/he cannot cut down on expenses for cooking fuel (food needs to be cooked), she would be left with 6.5 GEL for non-food expenses. If in addition she wants to store food in a refrigerator and pay the corresponding electricity bill, she would be left with a bit less than 3 GEL a month for all other expenses. These 3 GEL would need to be distributed among transportation, health care, personal items, etc.

Another option is to abandon the goal of a satisfying diet for a while and switch to less expensive food items in order to free resources for non-food items. This is a strategy that one would expect to see during the winter season primarily in Tbilisi, Kakheti, Samegrelo, Adjara and Javakheti where the poverty headcount can increase from summer to winter by 45%, 89%, 74%, 82%, and 90% respectively. The strategy is not free of risk because the possibility of illness increases not only in relation to the fall in diet quality, but also in relation to the length of time during which this trade off is maintained. The worrying findings of Table 2.2 (calorie intake by Georgian region) suggest that more than a few are already forced to trade off food consumption for other expenses.

Finally, let us repeat this exploration of options and opportunities for a family of four composed by two adults and two children with a total income of 115 GEL/month²⁰. If this family lives in Tbilisi, has a professional knowledge about nutrition, manages to find all products of the food basket and is able to capture all economies of scale, then the minimum budget for food would be 80 GEL. However, it should be noted that economies of scale in

food consumption is a crucial assumption, for without them the cost of food alone could climb to 140 GEL thus overcoming the entire family budget. If the breadwinner has to take a bus to work, then the disposable family income would be approximately about 27 GEL. If a refrigerator is used, somebody takes a quick shower once every four days, and only one light is kept on for five hours a day, the disposable income would then fall to 17 GEL. If atleast one room, only one for four people, is kept relatively warm and food is cooked, disposable income would fall to approximately 2 GEL. This leaves almost nothing for education in a family in which two children are of school age, nothing for health care, nothing for a bus ticket for the family member in charge of finding those good food deals, and nothing for personal hygiene. Here is when difficult choices have to be made. Where would you cut down? Would you consume less light in order to pay for school fees, or consume less food in order to pay for a visit to a doctor? This hypothetical family would show as "non-poor" in the Poverty Reduction and Economic Growth Program or in a poverty assessments in which the "Recommended Poverty Line" or "Alternative Minimum Poverty Line" is applied.

The next chapters will present a poverty assessment. In order to provide a broad picture, three poverty lines have been chosen to conduct this study. The poverty lines are defined for summer and winter periods and for each region of Georgia.

Table 2.5 shows three poverty lines. The first one is called the "NHDR Poverty Line". This poverty line results from using the NHDR food basket, prices reported from households in each region and the direct definition of non-food items as shown in Table 2.4²¹. The second poverty threshold has been denominated "NHDR Intermediate". It includes the food basket of the "Recommended Poverty Line" computed with prices reported by families in SDS surveys and the same non-food basket of the NHDR poverty line (35.4 GEL in winter and 26.92 GEL in summer for a single male adult). The third poverty line is called "NHDR extreme" and is simply 45% of the NHDR Poverty line. The three poverty lines in Table 2.7 capture the lower and upper boundaries of existing poverty lines while at the same time providing an idea about the choices open to individuals that have to make ends meet with them.

One of the principal objectives of this chapter has been to demystify poverty lines. Poverty lines are imaginary constructs that mainly serve the purpose of identifying groups that fall below this or that category. They provide shaky terrain for claims of "realism" or higher moral values as they merely reflect personal beliefs about what level of deprivation is incompatible with a particular definition of "poverty".

The NHDR has derived three poverty lines called "NHDR Poverty Line", "NHDR Intermediate Poverty Line" and "NHDR Extreme Poverty Line" and together they encompass the range of poverty thresholds currently being applied in Georgia. The value of the NHDR Poverty Line is close to the Official Poverty Line; the value of the NHDR Intermediate is close to the Revised Minimum Poverty Line while the vale of the NHDR Extreme is close to the value of the Alternative Minimum.

These poverty lines describe different types of poverty. The NHDR Poverty Line captures that segment of the population that earns enough to achieve a minimum diet but suffers non-trivial constraints in the provision of items like health care, education, electricity and gas. These are vulnerable families with little capacity to further restrict their budget for whom unexpected expenses can easily mean reductions in quality and quantity of food.

The NHDR Intermediate Poverty Line captures the group of families that are suffering non-trivial difficulties to achieve an adequate diet in addition to serious constraints in the provision of health and education services, electricity and heating. Though it is technically possible for these families to meet minimum calorie and nutrient intake, the task is not easy, for it demands a level of knowledge of nutrition that most Georgians appear to lack. It is likely that people having to live with the NHDR Intermediate Poverty Line are not achieving an adequate diet.

The NHDR Extreme Poverty Line captures the group of Georgians at the margin of society, a group of desperate people that are far from achieving an adequate diet and have almost no budget for health care, education, electricity and heating. This group can be thought of as people that are cold in winter and hungry all year round. They are probably physically debilitated due to the regular lack of enough food. For these Georgians, more than moderate illness can be a dangerous event and probably oblige them to sell whatever form of capital (tables, chairs, beds, etc.) still remains in the family.

The objective of working with three poverty lines is to facilitate the assessment of options and alternatives open to families. Together they provide a clearer understanding of the kind of life lived by those whom fall below a given poverty line, or in other words, a clearer picture of what it may mean to say, "50% of people in region X fall below the NHDR poverty line". For example, the 13% of poor identified in the Poverty Reduction and Economic Growth Programme have been categorized as such based on the "Alternative Minimum Poverty Line" (52 GEL). By comparing the Alternative Minimum to any of the NHDR poverty thresholds we can realize the kind of poor people we are talking about. This is a group worryingly close to the "NHDR extreme poor", a group of destitute and probably underfed Georgians almost completely cut off from health care and education services and highly likely to see a cycle of poverty perpetuated in their families.