1. THE COUNTRY

Colombia is a country with more than 40 million inhabitants, the third largest population in Latin America after Brazil and Mexico. It has an area of 1,139,000 square kilometers. its per capita income in 1997 was US$2,110, one of the lowest in South America, higher only than Paraguay, Ecuador, and Bolivia. On a world scale it is classified as a medium-low income country, occupying 67th place in 1993 according to the World Bank, without correcting for PPP as introduced by the United Nations.

Colombia has reached high levels of urbanization: 72% of the population is urban with an annual growth rate of 1.8%, equal to the average rate for Latin America and the Caribbean and 0.5% higher than that of the medium-low income group. The labor force is 27% agricultural and 23% industrial, and 37% are women, a higher percentage than that of Latin America, but lower than that of the medium-low income group.

From 1985 to 1994 the average annual rate of growth of per capita income in Colombia was 2.4%, a dynamic which, compared with that of other countries, could be considered acceptable, since the median growth rate for 120 countries in the same period was 0.8%.

Compared with the rest of Latin America, Colombia's performance has also been positive. In the last 25 years its GDP has grown at a higher rate than the regional average. During the 1980s (1981 -1990), which were especially difficult for Latin America, the cumulative variation of per capita GDP in Colombia was 17.9%. This positive result marks a clear contrast with the dynamic experienced by the region as a whole, where it decreased by 7.9% (sepal 1994, p.41).

In the 1990s Colombia begins to fall behind while the other countries show signs of recuperation. From 1991 to 1994, per capita GDP in the region grew by 6.1% and in Colombia by 8.6%. While Colombia continues to be above the regional average, the distance has been getting shorter.

2. VIOLENCE AND EQUITY

Violence in general, and more one with the persistence of the Colombian, is the result of a complex number of causes. Understanding the damages that the solidarity net has been suffered from violence in Colombia is a multidisciplinary work, and also suggests the solutions.

The objective is to contrast empirically the hypothesis that violence is a result of the combination of the following factors: the wealth, the inequality; the human capital endowment; the political participation, the municipalities expenses, and for the last period, the existence of army groups. The violence is measured by the rate of murders, the wealth is measured by the quality life index; the distribution, by the Gini coefficient, the human capital endowment by average years of education, the political participation by the rate of voter turnout in the mayor elections (1988 y1994), and the municipalities expenses by the transfers from the central government.

The data were taken from the municipalities in which exists comparable information. As quantitative tool a multivariable regression, of transversal cut for each one of the two periods (1985-1988: 19901996), where the variable for explain is the rate of homicides of the municipality.

The municipalities were divided in two groups in each period: the ones that increased their homicide rates in the period, and the ones which rates were equal or decreased.

For the total of the municipalities and for the municipalities with decreasing violence, the mean of homicides per 10.000 inhabitants stays practically the same in the 2 period (54 and 42 respectively). That means that the violence as a mean stays the same and does not increase. Never the less, the fact that the mean stays the same even though the maximal levels decreased, is an indicator the violence has been generalized to a level that was already high for a Latin American country.

The municipalities with increasing violence had, in 1993, a rate of homicides of 94 for 100.000 inhabitants, 36% higher than the mean of the ones in which the violence did not increased - 69 homicides for 100.000 - number that is still 3 times the Latin American mean. The municipalities with increasing violence had a QLI 5% higher that the ones with decreasing violence.

The model results suggest an interesting explanation of the persistence of the violence in the 10 years considered, when all factors are taken together. The analysis of the group with increasing violence tells us that when a municipality has a high level of wealth and its inequality levels are also high, there is a 30% probability that the municipality had increasing violence. This is true if the political participation and the mean levels of education are down. The government action does not moderate this tendency.

On the other hand, in the increasing violence group,it can be stated with 20% of probability that a 50% increase in the wealth does not increase violence. It is true if at the same time, the level of education grows in 10%, the inequality decreases at the same rate and the political participation increases. The increasing government expenses do not decrease violence.

The most important factor to increase violence in the 90 s was inequality. The data in the municipalities with increasing violence show the inequality importance as a social support of the violence. The accumulation of wealth effect is to maintain the same and the citizen participation decrease.

A homicide rate of 29 per 100.000 is not explained by the variation of the considered factors, and is more an effect of other variables.

The violence tends to reproduce with more probability in zones with high probabilities of wealth accumulation and with low government presence. The poor are not he cause but the victims of violence. This suggests a review of the extended idea that the violent zones must be a priority in the subsidy government expenditure. It is more important the institutional action to reach justice access and wealth accumulation possibilities.

The use of the power monopoly by the government to reach the private group disarmament is an important factor to reduce the violence in Colombia. There have also been identified two strategic elements that indicate possible solution ways: a higher level of education and effective democratic participation. A generalized increased of the educational level, in order to obtain a higher collective accumulation of human capital, must be an objective of any strategy to reach peace.

The necessary growth is not a fact of the historical tendency, but requires an special social mobilization that benefit the poorest population that have been excluded to increase the educational average and to corred inequality. In addition, the government action must change towards a better management, a more clear social information, and a more political seriousness and stability. This elements are important to create an effective participation and social control.

The democratic participation is another important element in the decrease of violence. The peace process, to be efficacious and permanent, must be centralized in the creation of collective democratic values that ensure the social active participation and not recognize the illegal use of the force.

It is a mobilization of all society components and an integral action between government and society. There must be always a new way of action and a clear and public management. It is required the existence of formal institutions because the government is more democratic, and informal institutions because the social participation build peace.

Dependent variable: Homicide tax 1993

Municipalities with increasing violence: the mean homicide rate increased between 1990-1992 and 1993-1995. Municipalities with decreasing violence: the mean of homicide rate don't increase.
The elasticities are the percentual variation of the homicide rate when there is 1% variation in the variable.
Notes:
* Homicide rate: Mauricio Rubio from the national army and Legal Medicince.
* QLI Quality of Life index, from Mision Social with municipal data
* GINI total, from Mision Social, with municipal data
* Education fomr Mision Social with census data
* Rate of veter turnout. From national Registraduria 1994
* Municipalities expenses per capita, national transferences from serires of DNP
Existence of army groups Statistical series of Alejandro Reyes.

3. THE HUMAN DEVELOPMENT INDEX

From the point of view of human development, Colombia is a country with big contrasts between regions and faces two important problems of great magnitude: inequality and violence. The level reached by the HDI changes when taking into account the effects of the violence and inequality

3.1 The HDI in Colombia

For 1997 the Human Development Index for Colombia is 0.768, putting it in 57th place on the world scale. Colombia is situated among the countries with "medium human development". In the Latin American context, Colombia shows an HDI lower than Chile, Argentina, Uruguay, Venezuela, Panama, and Mexico, and higher than all the others.

In the last decade, by either of the two measures, Colombia has shown continual progress in human development.

Table 1
Colombia: Human Development Index
1994-1997
 

YEAR   -- 1994 1995 1996 1997

LIFE EXPECTANCY AT BIRTH Value Index 69.3 0.738 69.8 0.747 70.2 0.753 70.7 0.762

EDUCATION RATES   Lieracy Schooling 90.6 64.0 90.9 65.2 91.1 70.1 91.4 72.0

EDUCATIONAL VARIABLES Index 0.817 0.825 0.841 0.849

ADJUSTED GDP PER CAPITA (PPP 1994) Value Index 5.575 .671 6.219 .689 6.028 .693 6.810 .699

HDI   0.742 0.753 0.762 0.770

Source: Report on Human Development in Colombia, 1999

The change in the HDI is directly related to the variations in each of the indicators of which it is composed, although not in a uniform manner. The largest effects are due to notable gains in per capita GDP and to advances in literacy and in the combined rates of schooling (primary, secondary, and higher education). Positive effects, though of a lesser magnitude, are observed in longevity, which is associated mainly with the reduction of infant mortality.

3.2 The difference between the city and the countryside

The first inequity form is between the urban and rural zones. In Colombia, the existence of huge social gaps between these two zones is clear, given that poverty particularly affects the countryside.

With the calculation of HDI by zone, the possibility is created of a measurement of the impact on human development of differentials in poverty or shortcomings in the basic capacities of the population. These disaggregated values measure the disparities which necessarily should be taken into account when the time comes to formulate policies or programs aimed at minimizing them.

Both the urban and rural populations have experienced significant gains in health, in reducing illiteracy, and in their ever more great participation in educational activities and in per capita GDP. However, there remain large disparities in the results achieved by the populations residing in each of the two zones.

Human development in both the urban and rural sectors shows a growth trend in the last decade. In the urban zone, the index reached 0.797 in 1997.0n the other hand, the human development of the population in the rural zone is still low, 0.699, and the gap in absolute values between the zones remains high, even if it decreased in 1997.

In this year life expectancy in the rural zone was ten years behind that of the urban zone. For the other two indicators the rural zone is even farther behind. Illiteracy is 4.1 times that of the urban zone, and in school attendance the rural zone would need more than 20 years to equal the urban zone if the speed of change of the last ten years is preserved.

Table 2 Colombia: Human Development Index by zone (urban and rural) 1985,1994, 1995, 1997
 
 

YEAR   -- 1985 1994 1995 1997

ZONE   -- Urban Rural Urban Rural Urban Rural Urban Rural

LIFE EXPECTANCY AT BIRTH Value Index 68.5 65.4 0.725 0.673 70.7 68.6 0.762 0.727 71.0 69.0 0.766 0.733 71.5 69.7 0.774 0.745

EDUCATION RATES   Illiteracy Schooling 7.05 23.06 58.6 37.8 5.57 19.59 70.6 47.0 5.42 18.85 71.0 48.3 4.70 19.4 77.2 59.7

EDUCATIONAL VARIABLES Index 0.815 0.639 0.865 0.693 0.867 0.702 0.893 0.736

ADJUSTED GDP PER CAPITA (PPP 1994) Value Index 4.576 2.423 0.713 0.449 6.481 3.544 0.696 0.595 6.866 3.854 0.706 0.610 7.682 1.029 0.725 0.617

HDI   UNDP 0.763 0.567 0.774 0.672 0.780 0.682 0.797 0.699

Source:Report on Human Development in Colombia, 1998 and 1999. Calculations by DIOGS, based on Census data, Household Surveys, and DANE (1997)

3.3 The Development Index in relation to gender

Human development implies the broadening of choices for all people and not just for one part of society. Development, if it does not include women, is unfair and inequitable. One of the most notable differences within the general HDI for any country is that which is found between men and women. It is commonly found that men are in better positions than women on almost all socioeconomic indicators except life expectancy.

The results found in Colombia demonstrate advancement toward equity between men and women in the last decade, with achievements in the areas of health, education, and economic participation. In 1985, the disparity in the conditions of the sexes was huge (GDI: 0.6467). By 1995, this disparity had been considerably reduced (GDI: 0.827), but it is still far from the desired equality. Given the greater achievements obtained in the development of their capabilities, a greater income share for women is a necessary condition for gender equality.

From the analysis of the factors which comprise the index (life expectancy at birth, illiteracy rate, school enrollment rate, income share) it is concluded that, rather than lacking basic human capabilities, women have been and continue to be economically discriminated.

Table 3
Colombia: Human Development Index with respect to Gender
1985, 1994-1995
 

YEAR   -- 1985 1994 1995 1996 1997

LIFE EXPECTANCY AT BIRTH Men Women 63.9 71.0 65.2 73.4 65.8 73.6 66.5 74.0 67.3 74.3

ILLITERACY RATE   Men Women 12.0 12.8 9.4 9.4 8.9 8.9 8.8 8.7 8.7 8.5

RATE OF SCHOOL ENROLLMENT Men Women 52.5 51.5 63.1 64.9 64.2 66.0 66.7 69.1 71.8 72.2

INCOME SHARE   Men Women 1.43 0.517 0.723 0.598 0.732 0.613 0.742 0.625 0.748 0.632

GDI   -- 0.646 0.718 0.727 0.738 0.748

Source: Report on Human Development in Colombia, 1998 and 1999. Calculations by DIOGS, based on Census data, Household Surveys, and DANE (1997)

Progress in health conditions and the effect of violence

Life expectancy reflects, in the case of Cc~ombia. not only health conditions but also, the prevailing situation of violence. While women increased their life expectancy by 2.6 years in the 1985-1995 period, for men it increased by only 1.9 years. This behavior indicates, on the one hand, that health conditions in the country have improved in general terms; but on the other hand, that violence has counteracted achievements in terms of lowering mortality and raising life expectancy, especially for men.

Violence particularly affects the male population, and of them the young adult population (25 to 50 years). While the number of violent deaths as a percentage of total deaths was 9.4% in 1985, by 1991 it had increased by 9 points to 18.4%, of which 92% correspond to male deaths. Beginning in 1992 the rate of violent deaths begins to decrease until reaching 15.5% in 1995. Thus, improvements in male life expectancy begin to be seen, even though the homicide rate is still very high.

Women more educated than men

The literacy rate of the adult population has improved in the last decade, but with more important progress in the female population, which in the 1985-1995 period reduced its rate by 4 percentage points as compared to only a 3 percentage point reduction for men. A notable result is that educational achievement - measured by the combined schooling rate - is higher for women than for men since 1989. In the reference period women increased their gross schooling rate by 14 points, while men do so by 3 points less (11 points).

If we assume equal educational opportunities by gender and equal assessments of the value of education, the explanation for the lower male gains must be sought in the differential migration from the countryside to the cities, with female predominance and greater male desertion to integrate into market activities.

Better economic opportunities for women

Although the female participation rate in Colombia, as in all countries, is lower than for males, it tends to grow over time. Real participation is higher than the measurement indicates, as it is a recognized fact that it is underestimated, especially in the rural zone.

In spite of the difficulties in estimating income shares, it is clear that women have obtained their greatest gains in the area of labor. Their share of income from work went from 26% in 1985 to 34% in 1995.

3.4 The HDI correctedfor income distribution

When corrections are made for income distribution, it is found that Colombia's human development index diminishes. Inequity in Colombia is equivalent to a retrocession of more than ten years of development.

Table 4
Colombia: Human Development Index corrected for income inequality
1994
 
 

	Educational Achievement Index	Life Expectancy Index	Adjusted per capita GDP Index	Gini	per capita GDP Index	HDI	HDI adjusted with Gini
1994	0.817	0.738	0.671	0.57	0.289	0.742	0.582
1995	0.825	0.747	0.689	0.56	0.303	0.753	0.593
1996	0.841	0.753	0.693	0.53	0.345	0.762	0.629
1997	0.849	0.762	0.721	0.56	0.340	0.770	0.630

Source: Report on human Development in Colombia, 1998 and 1999. Calculations by DIOGS, based on Census data, Household Surveys, and DANE (1997)

The importance of inequality can be observed when making projections of the evolution of the HDI in the next years with two choices. The first one, the inequality of income index is constant, with a Gini of 0.56. The second one, the Gini is decreasing until reaching in the year 2000 0.50. In the first case the HDI without adjustment go to 0.770 to 0.792, because of the improvement of the life expectance index (1.5 years), reductions of the illiteracy rate (16 percentage points), increase of the combined years of education from 0.713 to 0.80, without big modifications of the GDP. The inequality adjustment leave the HDI almost invariable between 1997 and 2000, in 0.66. In the second case, the HDI evolution after the inequality adjustment, allows an increase from 0.640 to 0.675.
 

Table No. 5
Colombia: Human Development Index, changing perspectives, according to different behavioral patrons of income inequality.1997 - 2002
 

Ano

Life expectancy rate

Illiterate population %

Combined schooling rate

Educational achievement index

Life expectancy index

Adjusted GDP Index

DHI 1997

Inequality adjustment Gini GDP with Gini Adjusted DHI by Gini

3.5 HDI and the cost of violence

The HDI allows, under certain assumptions, an estimate of the cost of violence in Colombia in terms of GDP. The effect of violence in terms of loss of human life has implied a decrease in the life expectancy of the population.

The question is now: what would life expectancy have been in Colombia if there had not been violence of the magnitude observed over the past decades? And, similarly, by how much would this lesser violence, transformed into higher life expectancy, have increased the HDI? Or to put it in other terms, by how much would GDP have to have risen to compensate for the losses in life expectancy and to reach the same HDI?

Violence particularly affects the male population, and of them the young adult population. Over the 1983-1995 period, more than 90% of deaths attributed to violence were of men.

The life expectancy gap by gender increases to 8.7 years in the 1990-1995 period (64.3 for men and 73.0 for women) with respect to the 1975-1980 period when the difference was only 4.5 years (61.8 years for men, 66.3 years for women).

In accordance with the performance of Latin American countries which, in the mid-1990s, achieved life expectancy ranges for women similar to those found in Colombia, the difference between genders would be between five and six years. If we accept as a valid assumption that, without the existence of heavy violence, a similar pattern would have been found, then male life expectancy would be from 67 to 68 years. This lets us conclude that Colombian men have lost three to four years of life expectancy (one and a half to two years for the whole population).

With average life expectancies higher by one and a half or two years for the whole population, the country would have achieved, in 1994, a Human Development Index 0.008 points higher for the former increase and 0.011 in the latter case.

Table 6
Colombia: Human Development Index, by value of life expectancy at birth
1994
 

Increase in life expectancy -- 1.5 years 2 years

Life expectancy at birth   Value Index 72.2 0.787 72.7 0.795

Education rates   Literacy Schooling 0.914 0.720 0.914 0.620

Educational variables   Index 0.849 0.849

GDP index   -- 0.70 0.70

HDI   -- 0.778 0.781

Source: Repcrt on Human Development in Colombia, 1998 and 1999. Calculations by DIOGS, based on Census data, Household Surveys, and DANE (1997)

To reach the highest HDI values, maintaining the life expectancy found for the country in 1994 (69.3 years), the other components of HDI would have to increase in a significant manner. In the case of GDP, the estimates indicate that its additional growth would have to be 16.1% to compensate for a loss of a year and a half of life expectancy, and 22 % to compensate for the loss of two years.¹

Table 7
Variations in GDP to compensate for losses in life expectancy
1994

Increase in life expectancy	Life expectancy at birth		Education rates		Educational variables	GDP Index	HDI
	Value	Index	Literacy	Schooling	Index
1.5 years	72.2	0.787	0.914	0.720	0.849	0.70	0.778
2 years	72.7	0.795	0.914	0.620	0.849	0.70	0.781

Source: Report or Colombia, 1998 and 1399. Calculations by DIOGS, based on Census data, Household Surveys, and DANE (1997)

3.6 Some conclusions

The HDI and the indicators from which it is composed show great progress in the country's human development for the period from 1985 to 1995. For 1995 Colombia has a similar position to what was found in 1994 for countries such as Panama, Venezuela, and Mexico, and is closer to the index of the developed countries.

The HDI, in both the urban and rural zones, has a growth trend associated with increases in per capita GDP, improvements in health conditions, higher literacy, and greater access to education. However, the socioeconomic gap between the country's urban and rural zones is reflected in the human development indexes. The indexes by zone are: 0.858 for the urban zone and 0.722 for the rural zone. Considering that the gains in average income go against the rural population, the gap can be expected to widen.

Over the last decade, gender disparities have been notably reduced. The GDI went from 0.646 in 1985 to 0.827 ten years later. In spite of gaining greater access and permanence in the educational system, which is reflected in higher schooling rates than for men, women were and continue to be discriminated against economically. A greater share of income and an egalitarian treatment in salaries for women still remain to be achieved.

Violence, measured by deaths by homicide, has diminished the life expectancy of the Colombian population, especially for young adult men. The cost of this violence can be measured by the additional effort the country must make, in terms of economic growth, to compensate for the reduction in life expectancy and to reach the same level of human development it would have if it lost less lives. The results of this exercise, show that the national GDP would have to have been between 16.1 % and 22.0% higher in 1997 for the country to have had the HDI which it could have achieved with a normal increase in life expectancy and no violence.

4. The Quality of Life Index

Building on the UNDP's Human development index, the Social Mission has developed the Quality of Life Index for Colombia. This index aggregates a measure of nominal income, human capital indicators, access to public goods and family composition. Plans are underway to incorporate measures of security and the enviroment as well.

The index incorporates a number of important advances. First, the use of statistical techniques that permits to incorporation of categorical variables and continues variables. Second, the section of the most powerful variables to differentiate between poor and non poor. The index is continues and statistically correlated with real income, as measured by household expenditures. Using administrative data, the index can be calculated for each municipalities in the country, which permits the measurements of the intensity and distribution of poverty.

Why the Index?

When evaluating the poverty of a person or of a region, there must be implicit a comparison of welfare. The measurement of poverty requires the identification of a group of people or regions that lack the goods and services contained in the predefined basket. In Colombia, the most used measurement of poverty is the Index of UBN (Unsatisfied Basic Needs). UBN measures basic physical condition of the household. The definition of the variable is fixed through time and some of the variables of such importance that their long-term absence will lead to early death.

The measurement of poverty with UBN does not permit the identification of how poor are the poor, neither which is the grade of inequality between them. The use of the UBN has been useful because it has help to introduce an alternative to the exclusive use of GIP and the economic growth as measures of welfare.

Another commonly used tool to measure poverty are the poverty and extreme poverty line, which is based on 1984 expenditure survey which has not been modify since then. This measurement has important limitations. It does not take into account no-monetary forms of income like homeproductions of barter, which are frequently used in rural economies. Likewise, in Colombia, changes in the structure of consumption are only captured every 10ryears with the new income and expenditure survey. In interim years these changes may be interpreted as change in the poverty level. Finally, the real income associated with government subsidies and services are not included. This is an important problem when we know that the implicit subsidy represent more than 60% of the income of the people in the poorest income decile.

Another argument that makes implicit the limitation of the income measurements, stated by A. Sen is that the income is viewed as the possibility to give capabilities to the households or to the families to reach desirable goods and services. The practice has shown, in many cases, that it is easily to look the deprivation of these goods than the income to reach them. At he same time Desai, has shown that the income concept, when it is used as a welfare measurement, has to be made as an ex ante estimation that searches to reach a consumers flow that the person can attempt maintaining its initial level of wealth. That means, that the objective is to measure the potential to reach a desirable group.

This way of looking the welfare as potentialities is nearest to Sen's theory of capabilities and functioning. Living is defined as the combination of beings and doings and quality of life is defined as the capacity of reaching this doings and socially valuable and reachable beings in time and space. Poverty is considered as failures or the "effective capacity" to reach a life standard which constitute the real privation or social exclusion.

A. Sen argues that the tile between goods and utility or satisfaction are complex and there are crucial differences to understand them... it is doubtful that utility is the last definition of life standard, even though it is interpreted as happiness, pleasure or wish satisfaction. The purpose is to concentrate in two intermediate states: the effective capabilities and the socially accepted groups.

In the actual Colombian situation, one of the socially accepted groups to be incorporated to the life standard are relevant variables as peace, justice and social capital. Even though these dimensions are good related with income, can not be measure through the personal income.

How was the QLI made?

The index combines in only one measurement the variables of potential access to physical goods, representing the physical house's characteristics with variables that measures human capital of the people in the household, the access possibilities to household public goods, the children and young's possibilities to reach school services and demographic characteristics to the households such as potential towards the labor market and the weight of children under 6 years. It is expected in the future to add the potential that adds environment, risk (citizenship security) and access to public goods to the household.

First Step: The variables are selected from a Socioeconomic Survey, taking in account what the theory and practice has found has more relationship with the population life standard. Using this kind of survey, helps to measure one moment in time, the principal aspects that help value the life conditions, with the same method and the same unit of observation (family).

Second Step: Each variable was defined so each observable situation with relation to it, can be classified for its contribution to the life standard. The problem of giving each category a different weight when the variables are not continuous, as house characteristics or the condition of assisting or not assisting to school, was solved through an statistic procedure for data analysis called "optimal quantification". This procedure gives numeric values to the variables categories so the relation between these observations and the measure objective is maximize. The procedure objective is to quantify the categories in such a way that hey assign the values taking into account all the variables together and controlling the quality of the results through statistical proofs on correlation and variance.

From the index logic's point of view this process allows an empirical verification of what Sen has called the selection of valuable objects. When considering all the variables in a group, it is assured that if you compare a certain combination A of valuable objects, with another combination B. if combination B has more than one object and at least the same amount of valuable objects, A has a higher life standard than B.

Third Step: Once the metric of the variables is unified, you must find the weight that each of the valuable objects has on the total standard: This was done in a series of steps:

1. The variables were grouped in a such a way that the logic's point of view was conceptually significative and from the statistics point of view, the variables included in a factor has the maximal relation between them and the minimal relation with those of another factor. The result was the grouping in four factors: a factor of physical potential that takes as an indicator the household's characteristics, a second one of potential access to the household's basic services, a third one of human capital and its potential over the labor market. Finally a fourth sociodemographic factor that takes into account the difference between the age structure of the family and the combination of people and physical resources that measure the "hacinamiento."

2.  A partial indicator for each of the founded factors was built. The indicator that was used is the principal component of the correspondent group of variables and, therefore, shows the contribution of each variable to the factor.

3. Later, the weighted mean that each of the four factors has, was found in a unique index, tor which it was found the first principal component of the four partial indicators. This measure allows the compensation between factors with a known average, which is a step forward in the consideration of having or not a quality. Finally, the scores were again standardized in a way that the indicator would be between 0 and 100 points.

The final result is a quality life index where each of the value objects has a weight and with which households can be classified depending on the value or category that each of the variables in the index has.

Once the variables and their weight had been identified, the households can be arranged. The index can be calculated with different data like the census, if departamental or municipal information is needed.

The index is considered as a step forward towards a better characterization and measurement of the life standards. It is needed now to identify the analytical characteristics having in mind that the index is a tool for practical programs and helps the conceptual solidity of its foundation.

MAIN RESULTS

The Quality Life Index is the result of the valuation in only one measure of 12 variables that measure the potential to reach the good and services' access and enjoyment in a medium term. The Index combines the household's potential generated by:

The result of the QLI weighted by size, poverty and inequality allowed to organized al the municipality. This organization of more than a thousand municipalities was divides into 20 groups, starting from the lower scores to the higher. The 20 groups show big difference in the main characteristics. The quality of life measured by the middle scores of the QLI is to 2.25 times greater in the group with a maximal score (group 20) than in the group with low score (group 1). The difference is somewhat greater between the scores of the variables of mean of the years of education of the head of the household and mean of the years of education of the people older than 12 years.

Taking as an approximation the years of education of the head of the household to value the gap between the municipalities of group 1 and 20, the first result is to note the distribution problem that exists in Colombia. There is a difference of 5 years of education between the first and the last ventil, that is to say 50 years calendar of development, given that in recent Colombian history 10 calendar years are required to increase the mean of years of education of the country in 1. This difference in education, that affects the potential of the households towards the labor market, is kept in the educational mean of the members of the households over 12 years.

To observe the level of scarcity, the years of education of the household's heads in 14 of the 20 groups, does not reach 5 years of basic and required education in Colombia since many years. None of the groups reach 9 years of basic mandatory education legally defined in the country since 1976. In the group that has the highest level of education of the head of the household, the scores show that not all of the children between 12 and 17 go to high school.

It can be noted that there is a small but positive advance. The attendance to primary school of children between 6 and 11 years is less discriminative. The means score of the highest group Shows that either all of the children attend to school in a household or the household does not have children that age. The group with the lower score is very close to the "all of them attend" category.

In house conditions, one of the main indicators in the frame of basic necessities the wall material, even though all the groups have a mean than 2.3 points which means material superior to "tapia pisada". The mean in more than half of the groups is closer to that material than to brick. The highest group has 5.7 points, which means that even here the construction with ceramic brick is not generalized. In floor materials, less than 4.3 points indicates, ground floor, broad wood or "tablon", which appears in 12 of the stratums.

In the access to public services the group of the 113 poorest municipalities (near 11% of the municipalities of the country and 2.4% of the households) do not have the minimal connections of water and sewage and are obliged to cook with wood or wood coal. For aqueduct services, 12 of the 20 groups have at least 4 points that is to say, they don't have access, not even to "carrotanque" or "pile publics". Only in the highest group the aqueduct supplies all the population.

The most serious problem is that of sanitary services, the poorest 5 groups do not have even "latrines but none of the groups have full coverage of sanitary establishments connected to a municipal net. The indicator of the materials with which they cook show that at least in half of the groups they use wood to cook in a percentage of families that affects the mean, (more than 25% of the households). The score lower than 4.8, with undesirables effects that these can bring to the ecosystem.

The differences in the room conditions are also very large between the richest and the poorest. The group with the highest mean has 2 persons per room, and the lowest 4 or more. One person more per household, that tends to be younger than 6 years is another difference between the poorest ventil and the richest. This difference is greater today than 20 years ago.
 
 

Notes:
¹With the another form of calculation for the GDP the change eas 2.5 % for compensate the lost of years in life expectancy