December. Working Paper PEP-CBMS. Review of Existing Monitoring Systems. Series

Transcription

1 Working Paper Review of Existing Monitoring Systems December 2011 PEP-CBMS Working Paper Series

2 RESEARCH PAPER 1 Review of Existing Monitoring Systems First Version, February 28 th 2011 This version, June 11 th 2011 Authors: Institute of Economics, UNICEN, Argentina

3 1. Introduction Historically, Argentina had low rates of unemployment. Until the late 1980 s it was close to 5%. The poverty level had a peak during the military government in the early 1980 s, but never went beyond 40%. There was, nevertheless, large heterogeneity in the country, with some provinces historically and structurally with very high poverty levels (located in the northern part of the country). 1 The structural changes in the economy in the 1990 s caused important changes in the labor market and in the socioeconomic conditions of society. The currency board (fixedexchange rate) together with sudden economic liberalization led to the closure of business that could not keep up with external competition; the reform of the state and its downsizing through privatization made thousands loose their jobs and the low cost of capital goods caused a replacement of work with capital. The unemployment and underemployment increased to unprecedented levels: in the worst part of the decade (during the Tequila crisis) it climbed up to 20%. The poverty decreased in the early 1990 s, because price stability (due to currency board ) improved at first the conditions of the population, which suffered the hyperinflation of 1989, but then it started to rise steadily and continued that way until In that year, the poverty levels exceeded, for the first time in history, the level of 50%. Since 2003 the economy has been growing fast, around 7% yearly, but inflation has reappeared in the country (currently around 25% annually) what has limited the growth effect on the poverty rates. There is also since 2007 controversy about official figures of inflation and poverty what has created a large debate in the country. For instance, official poverty rate is currently around 9%, but private estimates go from 22 to 30%. What is clear of this modern economic history of Argentina is that the country not only has not been able to achieve sustained economic growth, but it has created poverty. In particular, as Figure 1 shows, it seems to be the cases that each large crisis generates a new and higher level of poverty. The increase in poverty during the crisis is expected, but the concern in Argentina it has been the durable effect of the crisis, what some have attributed to the lack of safety nets. In fact, the system was almost inexistent until 1995 when the Plan Trabajar was created, but this was a badly designed and implemented program covering a small share of the poor population. 2 1 Poverty measured by income can be traced back to 1974 for Buenos Aires, which is not particularly one of the poorest regions of the country. To analyze early periods, the only source available are Census (the first in 1869), but with this information only it can be measured poverty by basic needs. Of the three standard ways of measuring poverty (by income (or expenditure), by basic needs or multifactor measures) Argentina only follows the first two. By income is measured with households surveys that provides information of the labor market for large urban areas, and by basic needs is compute for the entire country every 10 years based on the Census of Population and Housing. 2 Plan Trabajar I had beneficiaries and was the first of many social plans to come. It consisted on a small benefit for unemployed workers that perform community service.

4 Figure 1. Poverty and Unemployment in Argentina (Buenos Aires) 54% Hyperinflation Crisis 2001/ % 44% 44% Poverty Rate 34% 24% Poverty Rate 17.8% Tequila 27.1% 34% 24% Unemployment Rate 14% 10.6% 14% Unemployment 4% 4% Source: Author s elaboration based on INDEC data. Notes: Poverty corresponds to the income measure based on EPH. Both series were computed for the Great Buenos Aires Region (GBA), which comprises the city of Buenos Aires, the most important city in Argentina, with ¼ of the population living in GBA Nowadays coexist many social plans, from different public agencies (Ministry of Labor, Ministry of Social Development, Ministry of Education and Social Security Funds National Administration), in an uncoordinated attempt of fulfilling people s basic needs and to subsidize unemployed workers. To name some of them: PLAN JEFES DE HOGAR, addressed to unemployed workers with under-18s children. It consists of a benefit of the equivalent of approximately 40 US dollars monthly in exchange of community service. ASIGNACIÓN UNIVERSAL POR HIJO, addressed to unemployed workers, it consists of a benefit of approximately 50 US dollars monthly for every under-18s children in charge. PRESTACIÓN POR DESEMPLEO: addressed to workers that recently lost their job, it consists of a benefit of approximately between 60 and 100 US dollars monthly, for the following months after the lost of the job, decreasing progressively. PLAN NACIONAL DE SEGURIDAD ALIMENTARIA: addressed to those families in a poverty and nutritional vulnerability situation, it consists of the handing of food to families and the help to food self-production initiatives.

5 PROGRAMA NACIONAL DE BECAS ESTUDIANTILES: addressed to school students, it consists of a benefit of approximately 100 US dollars yearly to those students that need help for the expenses of studying. Critics of the social plans say that most of them are handed over with political criteria instead of an uncovered basic need criteria (in Latin America, it is called clientelism ), programs overlap and many people receive simultaneously help from different plans, there is not a system to identify potential beneficiaries, and neither there is control or systematic impact evaluation of the programs. Figure 2. Provinces according to Poverty by NBI, Census 2001 Up to 12% From 12.1 to 15% From 15.1 to 20% More than 20% Source: INDEC Notes: NBI poverty refers to poverty according to an index of basic needs that take into account 5 variables, if a household does not have one of them, it is considered poor. The Census 2010 was already implemented, but the final results were not available at the moment of writing this report

6 2. Institutions The main institution in Argentina responsible for statistics is INDEC (Instituto Nacional de Estadística y Censos Statistics and Census National Institute). It is the technical government agency responsible for the coordination and supervision of all public statistical activities taking place in the Argentine territory. Argentina, as a federal country, has statistics offices or secretaries at the three level of government, national, provincial and municipal 3. All these agencies are all integrated into and coordinated by the SEN (Sistema Estadístico Nacional - National Statistical System). In each of the 24 Argentine provinces operates a provincial Statistical Office (DPE, Dirección Provincial de Estadística), under the control of the respective provincial government. Each provincial Statistical Office coordinates statistical activities in its respective province and takes active part in collecting, entering and processing information at that level. The INDEC coordinates the SEN operations under the principle of procedural, methodological and standards' centralization coupled with operative decentralization. It draws up the Annual Program of Statistics and Censuses and develops the methods and standards that will ensure the comparability of statistical information originated in different sources. At municipal level, the rule is that there is no Statistical Office. There are some exceptions, although, and there is a renewed interest by statistics at the municipal level at the same time expenditure has been decentralized more in the last 20 years. Some municipalities have been creating Agencies or Secretaries, in fact there is a line of credit from the Interamerican Development Bank (IADB) financing this. 3. Poverty measures in Argentina Poverty by Income Level Poverty can be monitor in Argentina with: (i) Continuous Household Survey (CHS) or (ii) Census of Population and Housing. CHS, previously known as Permanent Household Survey (PHS) is a stratified sampling survey administrated in the main urban areas of the country, representing almost 80% of population. It started as a punctual survey, but since 2003 has been improved to be a 3 The basic structure of Argentina's government (rule by its 1853 Constitution and its Reform in 1994) is federal and republican. There are three levels of government: National Government, Provincial Government, and Municipal Government. The three levels are directly elected by the people. Currently, there are 24 provinces (in fact 23 provinces plus the autonomous city of Buenos Aires), and almost 1 thousand municipalities. Each level of Government has its own responsibilities and duties, and each level collects taxes.

7 continuous survey: during every quarter different fixed areas are surveyed per week (every area is surveyed a week per quarter, always in the same order). It has progressively incorporated new large and medium cities of the country. Nowadays it includes 31 urban centers. CHS is not designed to monitor poverty or living conditions, rather to monitor the labor market. With the income data collected the National Institute of Statistics (INDEC) produces a measure of poverty and extreme poverty as a headcount ratio below established line. The line takes into account demographic factors of the household (age and gender) but not economies of scales or other factors. The PHS was administrated twice a year, CHS is administrated continuously but databases are available twice a year as well. Databases with micro data are available in the website of INDEC. The earliest survey was administrated in 1974 in the Great Buenos Aires (GBA) Region, which comprises Buenos Aires city and surrounding areas. In the 1980s and 1990s sequentially more cities were added, since middle 1990s it is representative for the entire urban areas of the country. Technically with regard to poverty, the survey measures the income of the families, and compares it with the CBT (Canasta Básica Total, which is the amount of money necessary to fulfill personal basic needs) and the CBA (Canasta Básica Alimentaria, which is the amount of money necessary to fulfill a person energetic and proteic needs). The population that does not reaches the CBT and CBA income is considered under the line of poverty and extreme poverty respectively. 4 This income measure of poverty has the usual critics. In the case of Argentina the critics are: CB are based on the consumption pattern of the median family, which has been critized for nor reflecting the real situation for the poorest (overestimating consumption). CB are biased to the consumption patter of large cities, nor reflecting the reality of middle size cities. Income measure from Household survey only takes into account current income, therefore it is is influenced by income shocks Current income is just labor monetary income. It does not take into account nonmonetary private income, such as consumption of self production, fringe benefits provided by the employers, and imputed rents for owner-occupied accommodation. It does not include either a monetary measure of public transfers, neither for the income transfer nor for in-kind transfers (education subsidies, public health care, public housing, among others); Poverty by Unsatisfied Basic Needs 4 Technically what is computed is CBA as a basket of basic food. The Inverse of Engel convert CBA into CBT (this inverse relates total consumption to food consumption according to Consumption Surveys, which are administrated every 10 years).

8 Census of Population is administrated every 10 years since 1869 with some interruptions. Last Census was administrated in 2010 and definitive results are not available. The Census is administrated by INDEC in the entire country. Micro data is not publicly available. The measure of poverty computed from the Census is Poverty by Unsatisfied Basic Needs (NBI poverty by its Spanish initials) based on five pillars: a person is considered poor if it lives in a household having: (1) more than three persons per room (Crowding); (2) living in a house made of irregular materials, or in rented rooms (housing); (3) not having an indoor flush toilet; (4) having a child between 6 and 12 years that is not attending to school (school attendance) or (5) having four or more persons per person working and a household head with 2 or less years of primary school (subsistence capacity). A person is poor if one of the conditions is met. NBI poverty can be measured also with CHS, but only for large urban areas. Census has the attractiveness of representing an accurate picture for the entire country. But, in the way the indicator is defined, it is not influenced by income shocks, and it is not very dynamic. It is an important contribution to measure poverty when it is combined with indicators related to income. The main problem with this NBI index, like many, is that it is a mix of indicators without a clear purpose. Most of these indicators are inputs, rather than results, and some are proxies for income or consumption expenditures. The first two, crowding and housing conditions, are clearly related to household income, while having a flush toilet may be as well. Likewise, the fourth variable measures school attendance, but not achievement, and it is not clear whether the lack of school attendance is related to availability of services or not. Finally, having four of more persons per employed person, or a household head with only 2 years of primary education, is essentially a proxy for average household income, the presumption being that those with inferior educations and large households are spreading a small income over many people. Studies show that these indicators are not a good proxy for income. Arrieta and Caminos, for instance, find that crowding and sanitary conditions are the best proxies, while school attendance is rarely important (this is due to the fact that this variable indicates poverty when a household has a child aged 6-12 not enrolled in school, and in Argentina primary school enrollment is 98%). Correlation between the income and NBI measures is not perfect. For instance, in 1996 only one-third of the population which was poor by income was also poor by NBI. On the other hand, in the upper 60%, about 8% were deficient in terms of NBI. 4. Monitoring Systems SIEMPRO The first attempt to create a national level monitoring system of living conditions was SIEMPRO (Sistema de Información, Monitoreo y Evaluación de Programas Sociales or in

9 English System of Information, Monitoring and Evaluation of Social Programs). It was created in the middle 1990s by the CNCPS (Consejo Nacional de Coordinación de Politicas Sociales National Council of coordination of social policies, which reports to the President). Its objectives were producing microeconomic level information for policy formulation and impact evaluation. Also it was responsible for collecting administrative information on social programs in place, monitor the degree of progress in meeting the goals set in these programs, and assess the degree of correspondence between expected and actual impact achieved through the implementation of them. SIEMPRO with the financing and the advice of the World Bank designed and implemented a Living Condition Survey, similar to others applied in Latin-America in the same period. The first survey was administrated in 1997 and the second in 2001 with the financing help of the World Bank and IADB. It was expected another survey in 2006, this time financed by the Argentine government, but with the economic crisis of 2002 SIEMPRO team was dismantled, and the 2006 survey was never implemented. SIEMPRO was a dead agency during most of the 2000s but recently has been refunded with the objective of collecting information of beneficiaries of social assistance and create a list of potential beneficiary to organize better these programs (since it was found overlapping of the different programs favoring some families). Recent Changes in SIEMPRO SIEMPRO has been recently reopened to collect information of beneficiaries of social assistance programs, but the information has not been public. Within the sphere of SIEMPRO, it was created a Social Card (Ficha Social), in order to gather different beneficiaries of social plans into one single database. All government agencies and all government levels had to fill the Social Card in. Therefore, they received the software to process the information. The Computing unit of SIEMPRO, in addition to the processing software, has been acquiring georeferencing capability to specialize in the implementation of a Geographic Information System (GIS) All these process is in development. According to the local authorities interviewed for the present project, they filled the Social Cards, but they have no idea what happened with that information. They have not received any feedback from SIEMPRO, neither training courses nor software. The National System, therefore, is not really in place and it is not clear whether the recent changes are just cosmetics (after almost 10 years without paying attention to monitoring) or there is really a turning point.

10 Box 1.Objectives and organization of SIEMPRO The specific objectives of SIEMPRO are: To establish a System of information, evaluation and monitoring of national social programs. To develop and implement the SISFAM (System of Identification and selection of beneficiary families and Social Services). Strength national and provincial social areas in the development of monitoring systems and evaluations. To produce new information through the Social Development Survey. To ensure availability of necessary information on the population in poverty and social vulnerability and the implementation of social programs aimed at addressing it. To train public officers of national and provincial government areas in politics and social management. To disseminate and transfer methodologies and information systems to national and provincial agencies. To foster links between academia and the state in research and development of social policies and programs. SIEMPRO has the following areas: SISFAM: Its main objective is the identification and selection of families living in a poverty and social vulnerability situation through census methodology, in order to enable efficiency, transparency and fairness in the distribution of public resources. Its implementation by provinces and municipalities is voluntary. This area gives them training, technical assistance, software and financial support for initial surveys. Monitoring of Social Programs: Aims to survey, process and manage the information referred to implementation and execution of national and provincial social programs. It produces periodical reports about programs performance and relevant information about current provincial situations, including a diagnostic of political, social and economic situation. Social Program Evaluation: Performs techniques and methodologies to appreciate, in an objective way the results of public actions, in terms of rights protection, fairness, improving of quality of life and the extension of citizenship to the most vulnerable sectors. Social Analysis and Information: Provides information to other government agencies. For this, uses articulated instruments and different degrees of aggregation and update of social information. To build the system, social information basis from different sources (National Census; PHS; sector statistics; Education censuses; etcetera) were prepared. Computing: It aims to provide technical support to the program net, its servers and users terminals, as well as the design, development and maintenance of the software necessary to fill and supervise databases. This area is specialized in the implementation of a Geographic Information System (GIS). This tool makes possible the quick graphic comprehension of social information obtained. Besides, the area provides detailed maps, easing the fieldwork. In this area has been developed: IMS: Integrated Monitoring System. Software developed for monitoring of social programs. SCP: Social Card Processing. Software developed to charge social cards massively. The software is designed in different universities, to which technical support is given permanently.

11 Non Governmental System Poverty has gained importance in the political and academic agenda. There are several Institute and Research Centers from the private sector, NGO and private and public university analyzing poverty issues in Argentina. Perhaps CEDLAS from UNLP (Universidad Nacional de La Plata) is currently the most active and prolific research center in Argentina studying poverty and income inequality, but they do not collect information, rather do academic studies based on public databases. Another nongovernmental attempt is the Programa Observatorio de la Deuda Social (Social Debt Observatory program), of the UCA (Universidad Católica Argentina Argentine Catholic University): It was created in 2002, and its main objective is the study, assessment and monitoring of the state of human and social development in Argentina, as from national and international regulation parameters, taking into account the fact that non-enforcement of such rules constitutes a measure of the "Social Debt". This multidisciplinary team is formed by researchers and scholars (from the Argentine Catholic University itself as well as the National Council of Scientific and Technical Researchers) who carry out the manifold studies of national reality measured in terms of development achievements and shortcomings. The Social Debt Observatory Program of the UCA implements a national survey yearly - as from which mainly include indicators of human development and social integration (Argentine Social Debt Survey) on homes, towns and children living in urban areas of the country. Currently, within the project " The bicentennial Argentina: ", the poll surveys 5700 homes. From these surveys, the Observatorio de la Deuda Social has a primary data base with reliable social indicators, methodologically dependable and with widespread academic and political-institutional acknowledgment and application. None of these are really monitoring system collecting new information. Sub-national Government Monitoring System Each of the 24 provinces of Argentina has, with different degree of sophistication, programs and institutions collecting information to monitor poverty. Statistical Offices collaborates with INDEC to produce the CHS, but also they have their own surveys, which usually are targeted to cities which are not covered or not similar to the cities included in CHS. In general these studies are surveys following a stratified sampling technique. Sample size usually is rather small and there is no continuity. Social Census are not common. We focus now in the two provinces that we will focus the CBMS study: Salta and Buenos Aires.

12 Buenos Aires Buenos Aires Province, for instance (the most populated province in the country) has the HLS (Household and Labor Survey). This survey follows the methodology of PHS and is administrated by the Direction of Provincial Statistics (DPE). It is administrated to medium/small cities not covered by CHS. In addition to this survey, the Provincial Secretary of Culture and Education, carries out three times a year, and under the responsibility of every school authority, a survey (at the beginning, during and at the end of the school year) in every school. It also prepares (together with Dirección de Información Estadística Agency of statistic information) a map of educational establishments, relating them to social indicators geographically. The Ministry of Labor of Buenos Aires Province performs the EIMTM (Encuesta de Indicadores del Mercado de Trabajo en los Municipios Labor Market of Municipalities Indicators Survey), that consists of agreements between the province government and municipalities to survey the labor market conditions of the last ones. At Municipal level there are some attempts to organize the data and create poverty maps, but this so far has been done with Census Data (available only at the censal unit, around 400 household unit) or administrative data (register of beneficiaries). The municipality of Tandil, for instance, it has been working together with the University UNICEN to construct indicators and mapping for six years. Olavarria municipality is currently doing a similar work, integrating information of different administrative databases and trying to construct indicators at the household level. The Municipality of Tandil -in order to improve governance, to better identify the needs of the poors, to organize and administrate the existing programs, and to get the community involved- divided the city in 14 geographical regions or zones. The division was made with the help of the University UNICEN (through the Institute of Geography) and based on socioeconomic characteristics of the population. Each zone is planned to have a Community Center, which is the link between the Municipality and the community. Currently there are 10 Community Centers working of the 14 originally proposed (the four still not implemented are in the richest areas of the city). The creation of the geographical zones and the implementation of the Community Centers have been an initiative of the Municipality implemented with the help of the Institute of Geography of the University (UNICEN). The interaction between both has been close since Initially the Institute only created the division in 14 zones. Soon the Municipality understood the limitations of the division. At the beginning resources were distributed among CCs, but without a clear idea of the local needs. For instance the gas tubes that municipality distributes to poor households without access to the natural gas network were initially equally distributed among CCs. The Institute of Geography then gathered georeferenced information based on several variables (such as public service coverage, home fiscal valuations, Census data, etc.). This allowed the creation of a map, at census radius

13 (which covers around households) with the local needs and an index of Vulnerability. The Municipality started using these maps to allocate resources. At this point it is necessary to make clear that in general, Buenos Aires province s municipalities tend to use more advanced measures of poverty than the rest of the country, where, apart from the National Census of Population and Housing and the CHS (in inland s major cities), no other measure is performed (like the case of Urundel, Salta). Salta The provincial government of Salta, through its Bureau of Statistics (in spanish Dirección General de Estadísticas del Gobierno de la Provincia de Salta - DGE), implemented a geo-referenced map for its capital city (also named Salta) to identify the poor neighborhoods and their needs as Tandil did. Salta, however, has gone a step forward than Tandil and they implemented what they called a Social Census. The Social Census was created by the DGE to measure poverty and identify the needs of the poors, in order to address social assistance correctly. The Social Census combines the methodology of a Census with questions in the spirit of a SIEMPRO Living Condition Survey. 5 It consisted of a questionnaire of 82 questions and was implemented by the professional staff of the DGE in the poorest neighborhoods of the city of Salta. The results of the census where used to perform tha named geo-referenced map. The Social Census of Salta has been the attempt which is closest to CBMS in terms of collecting indicators, but very different in spirit. In Salta the Census was performed by professionals, without interaction with neighbors. It was costly compared to CBMS, what means it cannot be implemented very often. In addition, to apply this methodology in smaller cities and rural areas might prove to be extremely costly. This was one of the reasons, according to the Provincial Government of Salta, to explain their interest in CBMS. They think CBMS strategy can be an excellent tool to collect information in the rural municipalities, and at the same time to get the community involved, to increase human capital and empowerment in the local governments, and to instrument a formal mechanism to improve the interaction between DGE and municipalities. Conclusions o The existing systems to monitor poverty are only partially effective, because poverty by income level and by NBI are incomplete ways to measure it. o Currently, the efforts to survey poverty in its many dimensions had been only successful in very specific spots. At a national scale, the System is not really in place, and there is no certainty about its future. o Currently, neither there is no system implemented to measure the effectiveness of social programs, nor objective and precise criteria to aim them. 5 Preliminar results of the Social Census of Salta can be consulted at

14 Annex 1. Poverty Measures in Argentina Table - Poverty Gap by Province. NBI Criteria # of Households Poverty Rate % of households which its need is Lack of Health Lack of pipelinelack of Lack of telephone Insurance water Argentina ,3 60,2 50,8 17,6 71, CABA ,1 50,0 21,8 1,2 49,3 Partidos del GBA ,5 60,5 46,4 6,4 60,5 Resto de Buenos Aires ,5 53,9 41,2 8,3 60,8 Catamarca ,4 51,5 58,3 36,6 84,0 Córdoba ,1 56,0 42,1 8,7 66,7 Corrientes ,0 69,2 64,7 37,6 86,2 Chaco ,6 73,8 76,5 39,0 90,8 Chubut ,4 56,7 37,0 6,4 64,8 Entre Ríos ,7 59,3 46,1 16,8 79,5 Formosa ,0 72,0 81,3 50,6 90,4 Jujuy ,1 57,1 58,0 37,1 85,8 La Pampa ,2 57,1 29,1 6,7 61,7 La Rioja ,4 45,8 57,5 30,0 80,6 Mendoza ,1 62,3 45,2 15,7 68,8 Misiones ,5 67,6 70,5 15,1 88,3 Neuquén ,5 61,6 41,7 12,8 65,8 Rio Negro ,1 61,5 43,9 9,5 67,8 Salta ,5 67,1 64,7 41,7 88,1 San Juan ,3 57,5 52,0 30,7 75,2 San Luis ,0 62,0 51,2 18,7 74,1 Santa Cruz ,1 42,4 25,5 1,5 53,0 Santa Fe ,9 57,1 44,9 14,0 69,9 Santiago del Estero ,2 69,8 83,2 52,5 92,3 Tierra del Fuego ,5 42,5 15,8 0,3 38,1 Tucumán ,5 57,9 63,4 27,4 84,9 Source: INDEC

15 Table 3 - Evolution of poverty. NBI Criteria Province # of households # of Poor Poverty Households Rate # of households # of Poor Poverty Households Rate # of households # of Poor Poverty Households Rate vs vs vs Argentina 7,103,853 1,586, ,562,875 1,410, ,075,814 1,442, % -26% -36% Ciudad de Buenos Aires 918,758 67, ,223 69, ,024,540 72, % -5% -4% Buenos Aires 2,865, , ,409, , ,921, , % -26% -35% Catamarca 42,868 16, ,133 14, ,776 14, % -35% -51% Chaco 150,616 67, ,588 62, ,182 65, % -26% -38% Chubut 62,710 18, ,984 17, ,725 15, % -35% -55% Córdoba 610, , ,149 94, ,262 97, % -34% -43% Corrientes 140,198 56, ,041 48, ,957 54, % -34% -41% Entre Ríos 219,880 61, ,730 44, ,715 46, % -38% -47% Formosa 61,373 28, ,687 30, ,408 32, % -27% -40% Jujuy 86,266 38, ,800 35, ,631 37, % -25% -42% La Pampa 56,705 10, , ,661 8, % -36% -51% La Rioja 34,853 11, , ,390 11, % -25% -45% Mendoza 274,527 56, , ,418 53, % -25% -36% Misiones 129,028 50, , ,004 55, % -23% -40% Neuquén 52,909 17, , ,351 19, % -44% -54% Río Negro 92,741 30, , % -37% -51% Salta 137,318 58, , ,407 66, % -20% -35% San Juan 99,191 25, , ,902 21, % -34% -45% San Luis 50,911 14, , ,644 13, % -32% -53% Santa Cruz 26,449 6, , ,834 5, % -33% -56% Santa Fe 663, , , , , % -30% -41% Santiago del Estero 122,631 56, , ,201 46, % -27% -43% Tierra del Fuego 5,761 1, , ,816 4, % 0% -39% Tucumán 198,289 72, , ,787 63, % -33% -44% Source: Indec

16