CHAPTER 17 INFLATION AND ELDERLY BRAZILIANS* Marcelo Neri Fundação Getulio Vargas (FGV)/Instituto Brasileiro de Economia (IBRE)/ Centro de Políticas Sociais (CPS)/Escola de Pós-Graduação em Economia (EPGE) Salomão Quadros Fundação Getulio Vargas (FGV)/Instituto Brasileiro de Economia (IBRE) André Braz Fundação Getulio Vargas (FGV)/Instituto Brasileiro de Economia (IBRE) Vagner Ardeo Fundação Getulio Vargas (FGV)/Instituto Brasileiro de Economia (IBRE) 1 INTRODUCTION In recent years Brazilian legislation has created various affirmative policies aimed at elderly people. 1 Two important examples are related with the 1988 Constitution and with the 2003 Elderly People’s Statute. The Constitution guarantee the purchasing power of social security benefits, without, however, specifying any reference index. The Brazilian Economics Institute of the Getulio Vargas Foundation [Fundação Getulio Vargas (FGV)], following its long tradition of creating price indices, such as the General Price Index of Internal availability [Índice Geral de Preços de Disponibilidade Interna (IGP-DI)], has begun to research specific segments, such as elderly people. This article details the first steps towards a methodology to gauge the cost of living index developed for elderly Brazilians [Índice de Preços ao Consumidor (IPC-3I)], which will be published on a regular basis. The structure of the chapter is as follows. The introduction is the first section. The second describes the calculation of the specific index and compares its structure with that of the general consumer price index (IPC). The third section comparatively analyses the cost of living for elderly people. The fourth section deals with the real evolution of the purchasing * This chapter was translated from Portuguese to English by Eoin O’Neill and reviewed by the authors. 1. For a description of Brazilian policies for elderly people, see the chapter by Camarano and Pasinato in this book. Arq_26_Cap_17.pmd 6/10/2006, 13:12 539
32
Embed
INFLATION AND ELDERLY BRAZILIANS*€¦ · The Brazilian Economics Institute of the Getulio Vargas Foundation [Fundação Getulio Vargas (FGV)], following its long tradition of creating
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
CHAPTER 17
INFLATION AND ELDERLY BRAZILIANS*
Marcelo NeriFundação Getulio Vargas (FGV)/Instituto Brasileiro de Economia (IBRE)/
Centro de Políticas Sociais (CPS)/Escola de Pós-Graduação emEconomia (EPGE)
Salomão QuadrosFundação Getulio Vargas (FGV)/Instituto Brasileiro de Economia (IBRE)
André BrazFundação Getulio Vargas (FGV)/Instituto Brasileiro de Economia (IBRE)
Vagner ArdeoFundação Getulio Vargas (FGV)/Instituto Brasileiro de Economia (IBRE)
1 INTRODUCTION
In recent years Brazilian legislation has created various affirmative policies aimed atelderly people.1 Two important examples are related with the 1988 Constitutionand with the 2003 Elderly People’s Statute. The Constitution guarantee the purchasingpower of social security benefits, without, however, specifying any reference index.The Brazilian Economics Institute of the Getulio Vargas Foundation [FundaçãoGetulio Vargas (FGV)], following its long tradition of creating price indices, suchas the General Price Index of Internal availability [Índice Geral de Preços deDisponibilidade Interna (IGP-DI)], has begun to research specific segments, suchas elderly people.
This article details the first steps towards a methodology to gauge the cost ofliving index developed for elderly Brazilians [Índice de Preços ao Consumidor(IPC-3I)], which will be published on a regular basis. The structure of the chapteris as follows. The introduction is the first section. The second describes the calculationof the specific index and compares its structure with that of the general consumerprice index (IPC). The third section comparatively analyses the cost of living forelderly people. The fourth section deals with the real evolution of the purchasing
* This chapter was translated from Portuguese to English by Eoin O’Neill and reviewed by the authors.
1. For a description of Brazilian policies for elderly people, see the chapter by Camarano and Pasinato in this book.
power of the minimum wage, using the IPC and the IPC-3I. Finally, in the Appendix,the background, methodology and structure of IPC-3I are given in greater detail.
2 THE STRUCTURE OF CONSUMPTION IN OLD AGE
The “depersonalisation” of inflation does not harm macroeconomic policy. Afterall, if increases in prices are generalised, even if not quite simultaneously, they willfelt to some extent by everyone. Monetary control reverses the generalised characterof the increase. This is only part of the story. Some prices increase more thanothers, in a more prolonged way and even more permanently. From the influenceof the exchange rate on productivity increases, to the level of competition in eachmarket, and through the preferences of the consumer, many factors explain thedifferentiated dynamics of various prices collected.
In this way, diversified inflation arises. The expenditure of some social groupscan be concentrated on items whose prices, over relatively long periods, increasemore quickly than other items. Using the results obtained from the last familybudget survey [Pesquisa de Orçamentos Familiares (POF)], carried out in 2002and 2003, FGV is currently identifying certain groups and constructing specificinflation indices, including the group of elderly people. Also in relation to specialprice indices, FGV intends to study the child and teenage segment, to assist theimplementation of programmes such as Family Benefit (Bolsa Família).
The Consumer-Budget Survey allows to capture private demands andbehaviour in order to support actions and studies at various levels of governmentand of the civil society. A properly designed public policy shall be able to cover theareas not catered for in the private sphere—while the POF/FGV allows us todiscover these areas. We are not going to have a normative discussion here, ratherwe will just look as some descriptive elements that are useful to the debate, suchas, for example, the recently reopened debate about the convenience ofdisenttangling the basic social security benefit from the minimum wage.
The design of an inflation index for elderly people requires great care. Theconcept of budget used in this type of survey is a family one and not individuallybased, therefore, it is not possible to calculate the cost of living index for elderlypeople this way—except for those families composed only of elderly people.Although we consider this calculation interesting, we decided to use a definitionwe judged to be more representative in terms of public policy decisions: families
2. An alternative possibility would be to use the familiar per capita concept, common in the social literature on poverty. In other words,we could divide the expenses of each household by its number of members, and afterwards regroup them according to certain characteristics,such as elderly people or children. This would give greater internal consistency to the range of indices. Also relevant from the poverty lifeperspective is the calculation of the level—and not just the variation—in the cost of life of special groups.
Arq_26_Cap_17.pmd 6/10/2006, 13:12540
541INFLATION AND ELDERLY BRAZILIANS
composed of at least 50% of elderly people.2 The main difference is, without a doubt,expenditure on health and special care that are responsible for 15% of the budgetof these families, compared to 10.4% for the total set of families surveyed (Table 1).In the United States, a country that is known to spend a lot on the prevention andtreatment of illnesses, families as a whole spend around 5% of their family budgets onthis type of expenditure, according to the Bureau of Labor Statistics. Among thoseaged over 65, health costs rise to 11.5%, while only in the over 75 age group do theyreach the 15% reported by Brazilian families with a majority of members over 60.
Dividing our reference groups by income into those with income above andbelow 8 minimum wages, we can see once again, similar proportions (14.5% and15.1%, respectively). However, there are important differences in the compositionof health expenditure per income group: health services including doctors, hospitals,laboratories, and health insurance plans account for 8.5% of the budget of thehighest income group and 5.1% in the lowest income group. On the other hand,expenditure on general medicines accounts for 3% and 5.8% of the budget ofhigh income and low income families, respectively. Other health expenses do not varybetween income groups. In this way, irrespective of the operational questions,health service policies, such as the prohibition of the differentiation of health insurancecharges based on age established by the Elderly People’s Statute, have a lowerpotential for affecting poor elderly people than actions aimed at improving theoffer of medicine, such as the “popular pharmacy” or generic medicine programmes.
In relation to the composition of large consumption groups, the maindifferences in expenditure between families where at least 50% of members areelderly people and the total set of families can be found in the higher expenditureof the former group on: a) health and personal care costs (15% against 10.4%,respectively); b) expenditure on food (30.2% against 27.5%); c) miscellaneousexpenses (5.79% against 4.44%); and d) slightly higher housing costs (32% against31.8%). The opposite, i.e., a higher proportion of expenditure among the totalpopulation in comparison with elderly families can be found in the followinggroups of expenses:
a) Transport (7.85% against 11.72%), probably reflecting the lower demandsand elderly people’s entitlement to free public transport. The impact of urbanpublic transport is 50% higher for population as a whole than for elderly families.
b) Clothing (3.67% against 5.4%). In this item, elderly women spend morethan men, probably due to their greater longevity.
c) Education, reading and leisure (4.43% against 8.74%). There are importantdifferences in this item. The higher expenditure on education of the total population
(1.86% against 5.76%) reflects the long time spent to obtain a return on investmentin human capital. In the case of reading and leisure, similar proportions can befound, despite elderly people having more time for leisure, probably because elderlypeople enjoy half-price tickets to the cinema and the theatre (0.40% and 0.48%)and their greater free time.
The analysis of health can be complemented by the National HouseholdSample Survey [Pesquisa Nacional por Amostra de Domicílios (PNAD)] 1998/IBGE data. The proportion of families with access to health insurance is highestamong elderly people; 27% of elderly people and 24.4% in the population as awhole.3 The composition of health insurance plans is similar in relation tocomplementary exams (96% for both), while there is a slight difference in theproportion of those who have the right be hospitalised (95.5% against 93.6%). Inrelation to the use of health services, however, the proportions are very different:21.4% and 13%. This difference has an impact on the cost of private health insurance,with elderly people having to pay more: only 27% of elderly people paid less thanR$ 50 a month for health insurance, compared with 48% for the total population.In this way, the recently enacted prohibition of the differentiation of prices perage group in accordance with the elderly people’s statute will not be innocuous.
PNAD also allows the analysis of the incidence of health problems. As shownin Table 2, 10% of elderly people had been sick in bed in the two weeks before thesurvey, compared to 4% for the population as a whole. Chronic illnesses, such ashypertension, affect 45% of elderly people compared with 10.6% of the populationas a whole. Cardiac problems (19% and 3.9%) and depression (11.9% and 4.9%)can also be interpreted as affecting elderly people relatively more. The POF/FGVindex allows connections to be made between the state of health and the structureof medicine consumption per therapeutic class, as shown in Table 3.
Comparing families with at least 50% of elderly people and the totalpopulation, we can see high differences in the consumption of vasodilatations forarterial pressure (46% and 22.7%), as well as in the consumption of tranquillisersand anti-depressives (15.6% and 9.4%). The opposite relationship can be foundin other types of medicines, such as contraceptives and hormones (4% and 6.8%)dermatological medicines (0.97% and 3.7%), vitamins (1.3% and 4.4%), anti-infection medicines (0.96% and 7.2%) and painkillers and anti-thermins (4.5%and 9%). The flu vaccination campaign among elderly people may have contributedto a lower use of painkillers and anti-thermins.
3. For an analysis of elderly people’s health insurance by, see the chapter by Camarano and Pasinato, in this book.
Table 4 shows the variations in the IPC-3I4 and IPC-BR indices and theirimpact during the 1996-2004 period on the expenditure of the two study groupsin this research. We expect to shown here the immediate determinants of thedifference in inflation between the two groups. The differences found in theweighted structures were the result of an accumulated variation between May1996 and March 2004 of 92.15% and 74.6% for elderly families and other families
4. Based on 1,384 families covered by the POF in 2002/2003, with at least 50% of family members being 60 or over. For a betterdescription of the methodology of the two calculated indices, see the Appendix.
TABLE 3PARTICIPATION OF MEDICAL EXPENDITURE BY THERAPEUTIC CLASS IN BRAZILIAN FAMILYEXPENDITURE[%]
Total Pop. Families with at Least 50% Elderly
Anti-infection 7.16 0.96
Painkillers and Anti-thermins 8.95 4.54
Anti-inflammatory 11.27 9.17
Anti-flu and Cough Sedatives 3.73 1.26
Anti-allergic 4.69 2.65
Vasodilatations for Arterial Pressure 22.67 45.99
Tranquillisers and Anti-depressives 9.37 15.64
Hepatic Medicine (Liver) 0.90 2.47
Kidney Medicine (Renal) 5.49 2.66
Anti-acids 2.00 1.92
Tonics 2.11 0.84
Contraceptives and 6.81 4.01
Appetite Modifiers 1.52 1.39
Dermatological 3.66 0.97
Vitamins (Including Imported) 4.39 1.33
Homeopathic/Natural Medicines 2.91 2.42
Laxatives 0.73 0.81
Expectorant 1.65 0.54
Total 100.00 100.00
Source: DGD/Ibre/FGV based on POF/FGV.
Arq_26_Cap_17.pmd 6/10/2006, 13:12546
547INFLATION AND ELDERLY BRAZILIANS
TABLE 4VARIATION IN CONSUMER PRICE INDEX OF ELDERLY PEOPLE AND THE TOTAL POPULATION—MAY 1996/MARCH 2004
IPC - Variation (%) ImpactDescription
Elderly Total Elderly Total
IPC-General 92.15 74.56 100.00 100.00
Food 109.60 69.09 36.65 25.69
Food Stuffs 116.81 72.75 35.81 23.79
Eating Out 45.08 44.11 1.26 1.98
Housing 118.05 89.23 37.82 35.70
Rent and Charges 68.86 58.03 7.94 9.68
Public Residence Services 273.29 229.16 26.52 22.43
Furniture 70.21 45.71 0.17 0.65
Bed, Bath and Table Clothes 47.95 51.68 0.23 0.24
Electro-domestic Appliances and 29.53 9.56 0.48 0.26
Misc. Utensils 22.20 38.16 0.15 0.40
Conservation and Repair Items 90.25 83.78 3.42 3.72
Residence Services 89.67 77.67 3.59 2.66
Clothing 9.87 5.64 0.53 0.39
Clothes 6.04 0.14 0.22 0.01
Footwear 8.23 8.65 0.08 0.15
Clothing Accessories 14.45 18.39 0.06 0.09
Fabrics 45.36 41.23 0.10 0.10
Clothing Services 27.69 24.25 0.03 0.06
Health and Personal Care 51.89 62.32 9.24 8.73
Health Services 66.09 61.37 4.84 3.81
Medical and Dental Products 39.63 80.16 2.52 2.45
Personal Care 44.01 49.11 1.83 2.34
Education, Reading and Leisure 82.56 82.46 4.17 11.57
respectively. The main “villain” responsible for inflation in elderly people werehousing; responsible for 37.8% of total inflation and 118% for the study group.This was followed by food (36.7% of total inflation for families as a whole and110% for elderly families) and health (9.2% of total inflation for families as awhole and 52% of inflation for the study group). This final result contrasts withthe IPC-Total index, where education and transport had a lower impact thanhealth.
3 INFLATION FOR ELDERLY PEOPLE
The affects of the different consumption structures of the different age groups canbe summarised in terms of the inflation indices for the period. Figure 1 shows theevolution of monthly inflation since August 1994 according to the generalconsumer price index (IPC-BR) and the index referring to elderly families (IPC-3I).Due to the differences in the rate of inflation since stabilisation in 1994, it ispossible to visually analyse the results. This data is presented month by month inthe Appendix.
Figure 2 shows the ratio of the two indices in use since August 1994. Theyfall continuously showing the inflation for elderly people has surpassedsystematically that of the population as a whole. In the years included, the variationin the IPC-3I index exceeded that of IPC-BR by 18%. This is equivalent to anannual average difference of 1.8% in the post-stabilisation period.
(continuation)
IPC - Variation (%) ImpactDescription
Elderly Total Elderly Total
Reading 104.98 102.75 0.63 0.70
Leisure 70.15 69.66 1.66 2.87
Transport 111.17 99.01 8.71 14.56
Public Transport 161.70 177.74 4.56 9.61
Own Transport 82.44 68.59 4.14 6.38
Misc. Expenses 64.09 61.66 2.87 3.36
Alcoholic Drinks and Tobacco 49.59 50.70 0.91 1.54
Other Misc. Expenses 109.50 85.61 2.88 2.07
Source: DGD/Ibre/FGV.
Arq_26_Cap_17.pmd 6/10/2006, 13:12548
549INFLATION AND ELDERLY BRAZILIANS
4 THE REAL MINIMUM WAGE
This section looks at the evolution of the purchasing power of elderly people,beginning by analysing the purchasing power of the minimum wage, the basicreference unit for social security benefits. Figure 3 complements the analysis inthe previous section tracing the trajectory of the absolute values in the two priceindices since August 1994, when it is possible to describe the levels of these series.5
As can been seen, the two gradually and continually drift apart and the IPC forelderly people rises above the IPC-BR. The differences in the consumer priceindices observed imply a relatively small increase in the purchasing power of the
5. The successive changes in currency that occurred during the launching of stabilisation plans makes the calculation of purchasing powerduring inflationary transitions difficult, especially in the case of the 1994 Real Plan, due to the adoption of the URV [Neri (1995)].
minimum wage during the period—when deflated by IPC-3I—in comparisonwith IPC-BR. In the first case, the minimum wage had a real increase of 33.1%,from August 1994 to May 2004. The increase rises to 53.8% if deflated by theIPC-BR. The different trajectories of the minimum wage purchasing power indexare shown in Figure 4.
In the 12-month-period ending in May 2004, inflation for elderly peoplewas 5.67% compared to 5.29% in the IPC index for the population in general.Although inflation for elderly people is at a higher level, its value is still lower thanthe variation conceded to the minimum wage, representing a real increase at theend of 2.52%. In this way, the substitution of the general consumer price index bya specific index that used the consumption structure of elderly people as a referencefor social security would have had a more favourable impact for this segment inthe recent period.
By way of conclusion, the results show that the inflation indices for elderlypeople allow a more precise measurement of the evolution of their effectivepurchasing power.6 They act as an alternative reference system to implement healthand social security policies. Their correct use, which is only beginning, is ournewest challenge. They also show that, due to the consumption structure of elderlypeople, they have been more effected by inflation in the last ten years that thepopulation as a whole.
6. Analyses at the purchasing power level, instead of the variation in purchasing power, require similar treatment to that observed in theliterature about poverty lines. The question of economies of scale, in particular, can play an important role in the differentiation of thepurchasing power at a given level of income among elderly families from others, due to the lower size of the forme [Ferreira, Lanjouw andNeri (2003)].
Arq_26_Cap_17.pmd 6/10/2006, 13:12550
551INFLATION AND ELDERLY BRAZILIANS
APPENDIX
IPC-3I
A.1 Background
POF aims to collect representative information on the expenditure and earningsof families in relation to the formation of consumer prices indices. The POF,whose aim was to define the weightings for the general IPC, was began in June1956. The survey was based on the completion of “household notebooks”, inwhich the type and value of different items of expenditure were entered on a dailybasis.
In the second quarter of 1966, the IPC of Rio de Janeiro used items andweightings from the POF carried out between 1961 and 1963. The weightingsreflected the consumption patterns of families earning less than five minimumwages, representing more than 2/3 of the population of Rio de Janeiro at the time.In 1972, the results of the POF (1966/1967) were introduced into the IPC, andthe weighting now covered 411 items of expenditure.
In January 1989, IPC began to be calculated for São Paulo, as well as Rio deJaneiro, with the structures being extracted for the POF carried out in 1986-1987. This strengthened the accuracy of the survey, with the cost structure nowcovering families with earnings between one and 33 minimum wages.
In 1992/1993 and 1997/1998, two other surveys were carried out. The priceand consumption indices derived from these surveys were used in the methodologicalchanges occurred in January 1994 and January 1999, respectively. In January
2001 the Getulio Vargas Foundation’s IPC index became even more representative,through its expansion to a further ten state capitals: Belém, Belo Horizonte, Brasília,Curitiba, Florianópolis, Fortaleza, Goiânia, Porto Alegre, Salvador and Recife.
The POF carried out by the Getulio Vargas Foundation (POF/FGV) in2002/2003 was used to prepare the IPC-3I. The population focused on consistedof urban families residing in Belém, Curitiba, Florianópolis, Fortaleza, Goiânia,Porto Alegre, Rio de Janeiro, Recife, Salvador and São Paulo. Members of thearmed forces and other quasi-military institutions residing in barracks andinstitutional housing, people in hospital, patients of various institutions, and peopleresiding in collective housing were excluded.
A random sample is made on four different dates, so that consumptioninformation can provide an updated empirical basis in order to allow thepreparation of average consumption structures. In this way, baskets of goods andservices can be classified. In each city eight field surveys were carried out—fourprice surveys and four family budget surveys.
A.2 IPC-3I
IPC-3I, as it is calculated, does not just cover families consisting solely of elderlypeople (defined as individuals aged 60 or over). The structure of the index wasproduced by families with at least 50% of their members being elderly people andwhose income is between one and 33 minimum wages.
The final basket of the index for elderly people has a similar structure to thatused by the traditional IPC published by FGV. The consumption structure isbroken down into seven groups, 25 subgroups, 88 items and 450 sub-items. Themain difference between the traditional IPC and its version for elderly people liesat the sub-item level, the traditional version has 25 more sub-items than the versionfor elderly people.
The most important differences can be found in the weightings used in theconsumption baskets. In the development of this index, important differences inthe consumption structure of elderly families were discovered. These differencescan be perceived through the weighting structure, used to reveal where theexpenditure of these family units is most concentrated. Table A1 shows the weightof the seven classes of expenses in the structure of the IPC-BR vis-à-vis that of theIPC-3I. Table A2 shows the weighting of the large groups of expenses for alternativemodalities of families that contain elderly people, such as those composed ofelderly people in general and where the family only has elderly people. A high rate
Arq_26_Cap_17.pmd 6/10/2006, 13:12552
553INFLATION AND ELDERLY BRAZILIANS
of overlapping was observed: 82% of people in elderly families were elderly people,corresponding to 1.5 elderly people per household. We highlight below thedifferences between the different groups of expenses analysed for the differentuniverses of analysis.
IPC-3I already has a historical series—lasting to May 2004. The firstcalculations introduced to the inflation indices the differences found in theweightings structures with total accumulated variations in the period from May1996 to March 2004 coming to 92.15% and 74.6%, respectively. Table A3 showsthe evolution of the structure of variations in IPC–3I and in IPC–BR and theirimpact during this period.
BIBLIOGRAPHYALMEIDA, A. N. de. Determinantes do consumo de famílias com idosos e sem idosos com base na
Pesquisa de Orçamentos Familiares 1995/1996. Piracicaba, SP, set. 2002 (Dissertação deMestrado da USP).
CAMARANO, A. A. (org.). Muito além dos 60: os novos idosos brasileiros. Rio de Janeiro: IPEA,dez. 1999.
HOBIJN, B., LAGAKOS, D. Social security and the consumer price index for the elderly. Currentissues in economics and finance, Federal Reserve Bank of NewYork, v. 9, n. 5, May 2003.
FERREIRA, F., LANJOUW, P., NERI, M. A robust poverty profile. Revista Brasileira de Economia,v. 57, n. 2, p. 59-92. Rio de Janeiro: Fundação Getulio Vargas, jan.-mar. 2003.
NERI, M. Sobre a mensuração dos salários reais em alta inflação. Pesquisa e Planejamento Econômico,Rio de Janeiro, v. 25, p. 497-526, dez. 1995.