Demographic changes and intergenerational transfers in Portugal Follow-up report: an analysis for 2010

This project aims to carry out an analysis of the effects of aging on the economy in Portugal. The application of this analysis to Portugal is particularly interesting given thehigh degree of population ageing and because it would allow to observe the way different policy measures influence the distribution of resources among different age
groups.

From this analysis, valuable lessons can be learned to understand the determinants of the well-being of individuals, especially those dependent on other age groups. The analysis is also useful to study the sustainability of the welfare state in different economic situations and how public policies can respond to demographic challenges.

This report covers the intergenerational dynamic referring to the year 2010 for Portugal. Afterwards, the analysis exposed here will be reproduced for other periods so one can observe the consolidation of the welfare state in Portugal throughout time, along with the effects of demographic and educational transitions.

2. The NTA methodology

The National Transfers Accounts (NTA) approach was developed in the early 2000s following an international project led by the University of Berkeley and the University of Hawaii. Currently, more than forty countries around the world participate in this project. The NTA estimation method has been approved and published in a United Nations (Population Division) manual. NTA consist of estimating, for each moment of time and for a given economy, all the flows of resources that take place among the different age groups of the generations that live contemporaneously. In economic terms, people's life cycle can be broadly
divided into three major stages: early childhood-youth, active age and retirement age.

Intergenerational transfers occur because the relation between how much people produce and how much they consume changes along these stages. When individuals consume more than they produce (childhood and old age), they need cash flows from productive age groups in order to maintain their consumption level. These flows are sustained by governments (public transfers such as education, health care or pensions), by families (private transfers between mainly family members inside the same household or in different households), or through capital markets (savings, borrowed loans, interest payment or reception, asset returns). Families, the public sector and markets are therefore the intergenerational redistribution mechanisms that allow the necessary transfers for children and the elderly to consume and meet their needs.

The difference, for each age, between labour income and consumption is the life cycle deficit (LCD). Inevitably, people have to face a deficit during the stages in which they do not have the capacity to generate income from work (childhood and old age), while for a good part of their active life they will generate a surplus (consumption lower than the labour income). The LCD must be financed through the three intergenerational transfer mechanisms mentioned above: family transfers, public transfers or reallocations of assets through the markets themselves. The importance of these three mechanisms is different in each country and has most likely undergone significant variations throughout history. The goal of this project is precisely the analysis of this temporal evolution, for Portugal, studying the interaction between the three mechanisms of intergenerational redistribution of income.

The procedure for estimating intergenerational transfers is complex and demanding in terms of searching for and processing statistical data at the micro level. The NTA provides not only age profiles of consumers and labour income, but also age profiles of all variables into which they can be broken down, as well as the different mechanisms for financing consumption needs. All profiles are obtained on a per capita and aggregate level (by multiplying each profile by the population in each age group). The aggregates must coincide with those provided by the National Accounts of each country, so that the ATN are consistent with National Accounts. 

3. NTA estimates: analysis of available data sources  

NTA estimation an intensive and thorough task, due to the large amount of data that needs to be analyzed. On the one hand, microdata on income and consumption are necessary to determine its distribution by age. For Portugal, income microdata is taken from the European Union Statistics on Income and Living Conditions (EU- SILC), while consumption data is taken from the Household Budget Survey (HBS). 

The Household Budget Survey (HBS) in Portugal has been carried out by the Statistics Institute since the 1960s. The first survey was that of 1967/68. The 2010/2011 survey was the first to make use of electronic recording of the daily consumption of goods and services. Consumption is broken down according to the Classification of Individual Consumption by Purpose (COICOP), which amounts to almost 14,000 products. 

The EU Statistics on Income and Living Conditions (EU-SILC) is an annual survey, also conducted by the Portuguese Institute of Statistics since 2004. The aim of this type of survey is to produce statistics on the income of individuals and families living in EU countries and to measure and compare poverty and social exclusion between countries. It has come to replace the former European Community Household Panel. In the case of Portugal, the sample is selected from the data of the 2001 Population and Housing Census (Censo 2001). 

Some social aggregates such as collective housing (hotels, hostels, etc.) as well as housing for social support, education, military, prison, religious, health, and others were excluded, which means that they are not part of the EU-SILC sample. The data collection for 2010 took place between May and July of the same year. The information was collected by direct interviewing individuals using a personal computer (Computer-Assisted Personal Interview) (INE, 2011). 

This project estimates the detailed ATN for 2010, which will then be extended for other periods for which sufficient data are available to create the age profiles for consumption and income, while estimating resource flows across ages and consumption patterns for public goods. Developments will be analysed. 

Income is broken down into several categories. In particular, it is necessary to distinguish between labour income (wages and salaries from self-employment, including unpaid work in family businesses), housing rentals (both real and imputed), capital income, private transfer income (intra and interhousehold) and public income (pensions, unemployment insurance, etc.). Labour income is that which is used to subtract from consumption to obtain the LCD. It is obtained on the basis of EU-SILC data. 

In order to identify resource flows across different ages, consumption data should be broken down into education, health and other categories, and for all of them, a distinction should be made between public and private consumption. 

Information on private health consumption is available in the FBS itself, while the public component is not collected there. The research team is seeking the information from either the Working Party on Ageing (AWG) or the OECD. So far, it has not been possible to access this information - that is why this report does not, as yet, analyze any NTA variables that relate to public health consumption. 

With regard to education, the EPF contains information on the educational level of each individual, as well as their private educational consumption in the reference year. To obtain public consumption of education, the information provided by the OECD and Eurostat is combined. From the former, it is possible to obtain the number of students by level of education, and from this, the amount spent on each level. With this information, it is finally possible to obtain profiles of public consumption of education, by age. 

Public consumption other than health and education (basically public transfer of income in kind) is also estimated on the basis of EU-SILC data. The Eurostat database is also used to assess information on the population by age and gender, as well as to calculate aggregate controls. 

4. NTA estimates: preliminary results for 2010

Here are some preliminary results of the NTA estimates for 2010. Figure 1 shows the age profile of per capita labour income. As expected, labour income is clearly concentrated in active ages. What is particularly interesting in the Portuguese case is the late peak in labour income immediately after age 50, which is not observed in many other European countries. See, for example, the age profile of labour income in Sweden in Figure 3.

If we disaggregate the age profiles of labour income in Portugal by gender (Figure 2), it is clear that the peak after 50 is driven by male workers, leading to a widening of the income gap between the sexes around this age. Another peculiar feature is how the labour income gap is basically non-existent for inexperienced workers (up to the age of 27), unlike, for example, Sweden, where it is present from early ages (Figure 3). The pattern of labour income is reasonably flat for Portuguese women in the age range of about 27 to about 49 years. 

Figures 2 and 3 also reveal that while men and women stop earning at similar ages in Sweden, the same is not true in Portugal. The income from work reaches the bottom by Portuguese women workers are around 70 years old, while for men this only happens around 80 years old. 

Private transfers between households were also estimated. These transfer inflows and outflows capture only financial transfers and are directly estimated using the age- and gender-specific averages of the EU-SILC variables and adjusted to match the net private transfers from the rest of the world (ROW) (Istenič et al., 2016). Figure 4 shows the gender differences in inter-household transfers for Portugal in 2010.  

This type of transfer takes place between households and includes direct transfers between households (e.g., alimony payments and gifts) as well as indirect domestic transfers mediated by non-profit institutions serving households (e.g., donations), but excludes capital transfers as legacies. The difference between the inter-household transfer and the inward and outward flows are private net transfers to/from the ROW (UN, 2013). It is important to note that the NTA methodology assumes that all inter-household transfers of income occur between each household head, as there are no detailed data on who exactly is responsible for the outflows or inflows. This is why the network transfers are basically zero for children, because they are never the head of the household. 

Net transfers are inputs minus outputs. In the case of Portugal, the estimate of net interbank transfers, household transfers are positive up to age 60, and again after age 76. Only between these ages do people give more than they receive in these exchanges between households. 

Once again, the gender differential is striking, as men have negative net flows throughout more than half their adult lives, while female heads of household are net recipients for most of their lives. 

The pattern of net transfers between households in Portugal is clearly different from the patterns in countries such as Germany or Greece (Figure 5). As can be seen, the German population has negative net transfers due to large private net transfers, while Greek individuals receive a significant amount of resources, particularly as young adults.

Some calculations of public transfers were also made. However, due to the above-mentioned lack of availability of an age profile for public health expenditures, the age profile of total public transfers is also not available. The age profiles for retirement pensions and unemployment benefits are presented here. 

Pension entries are recorded from the age of 45 (Figure 6). For people who retire considerably early (before age 60), there is no systematic difference between the values received by men and women. After that, pension benefits for women are always lower than for men. In addition to the difference in level, there is also a difference in qualitative form. Before the age of 70, men's pensions start to decrease, while this does not happen with women's pensions. 

Unemployment benefits, received in the active ages, increase equally for men and women up to the age of 30, which expresses the very low labour income gap we see in Figure 2, but also similar levels of unemployed persons. From the age of 30 onwards, the age patterns of unemployment benefits are very different, with a peak for women at around 40, but much later, in the 53-57 age range, for men.

In fact, when looking at the number of unemployed people by age group in 2010, in Portugal, the age group of 55 and over is the only one with more men than women in unemployment. This adds to the higher levels of labour income of men than women, particularly pronounced in the 5th age decade. 

5. 5. Next objectives 

The next steps of our project are as follows: 

- Determine the age profile to be used in public health consumption and consequently calculate the LCD; 
- Collect more data and estimate the ATN for other years/periods; 
- Detailed analysis of the results in coordination with the rest of the sub-projects 

References 

Abio, G., C. Patxot, E. Rentería and G. Souto (2015). “Taking care of our elderly and our children. Towards a balanced welfare state”, in M. Gas-Aixendri and R. Cavalloti, Family and Sustainable Development, Thomson Reuters, pp. 57-71.

Albertini, M. (2016). “Ageing and Family Solidarity in Europe: Patterns and Driving Factors of Intergenerational Support”. Policy Research Working Paper 7678, Poverty and Equity Global Practice Group, World Bank Group.

Albuquerque, P. C. (2014). “Intergenerational Private Transfers: Portugal in the European context”. European Journal of Ageing, 11(4), 301-312.

Altonji, J., F. Hayashi, L. Kotlikoff (1992). “Is the extended family altruistically linked? Direct tests using micro data”. American Economic Review, No. 82:1177–1198.

Auerbach, A. J., J. Braga de Macedo, J. Braz, L. J. Kotlikoff and J. Walliser (1999). “Generational accounting in Portugal”, in Auerbach, A. J., L. J. Kotlikoff and W. Leibfritz (editors), Generational accounting around the world. The National Beareau of Economic Research.

Bloom, D.E. and J.G. Williamson (1998). “Demographic Transitions and Economic Miracles in Emerging Asia”, The World Bank Economic Review, 12(3), 340-375.
Esping-Andersen, G. (2002a). “Why we need a new Welfare State”. Oxford University
Press.
INE. (2012). Inquérito às despesas das famílias 2010/2011.
INE. (2011). Inquérito às condições de vida e rendimento 2010/2011.
INE, IP. (2009). "Projecções de população residente em Portugal 2008-2060."
Destaque, Lisboa-Portugal.

Istenič, T., Hammer, B., Šeme, A., Lotrič Dolinar, A., & Sambt, J. (2016). “European National Transfer Accounts”. Available at: http://www.wittgensteincentre.org/ntadata.

Lee, R. and A. Mason (2011). “Population Aging and the Generational Economy. A Global Perspective”, Edward Elgar.

Mason, A. and R. Lee (2011). “Population aging and the generational economy: Key findings”, in A. Mason (Eds), Population Aging and the Generational Economy. A Global Perspective, Edward Elgar. OECD (2014). “Society at a Glance 2014: OECD Social Indicators”, OECD Publishing. http://dx.doi.org/10.1787/soc_glance-2014-en.

Patxot, C., E. Rentería and G. Souto (2015). “Can we keep the pre-crisis living standards? An analysis based on NTA profiles in Spain”, Journal of Economics of Ageing, 5, pp. 54-62.

Patxot, C., E. Rentería, M. Sánchez-Romero and G. Souto (2011). “How intergenerational transfers finance the lifecycle deficit in Spain”, in R. Lee and A.Mason, Population Aging and the Generational Economy. A Global Perspective, Edward
Elgar, p.241-255.

Patxot, C., E. Rentería, M. Sánchez-Romero and G. Souto (2012). “Measuring the balance of government intervention on forward and backward family transfers using NTA estimates: the modified Lee arrows”, International Tax and Public Finance, 19, p.442-461.

Pinheiro, J. (2018). “Generational Accounting for Portugal”. (Unpublished master dissertation). Universidade Católica Portuguesa, Lisbon, Portugal.

Rentería, E., G. Souto, I. Mejía-Guevara and C. Patxot (2016). “The effects of education on the demographic dividend”, Population and Development Review, 42 (4),p. 651-671.

Sánchez-Romero, M., G. Abio, C. Patxot, G. Souto (2018). “Contribution of Demography to Economic Growth”, SERIEs Journal 9, pp. 29-64.Tiefensee, A., & Westermeier, C. (2016). Intergenerational transfers and wealth in the Euro-area: The relevance of inheritances and gifts in absolute and relative terms. UN (2013). National Transfer Accounts Manual. Measuring and Analysing the Generational Economy. Population Division, Department of Economic and Social Affairs,
United Nations, New York.