Taking age into account
Published 18 Apr 2017
Treasury Staff Insights: Rangitaki article by Paul Rodway
I sometimes wonder if my two kids, born in the late-1980s, will be part of a generation that won’t have the economic resources, income and assets, in their middle age as my generation had. There is a sense that compared with the Baby Boomers, many of these Gen Ys are closing out their 20s with high student debt and only moderate incomes, and can’t see how they could ever own a house. And while not all Baby Boomers are facing an easy retirement, many have savings, are living in largely mortgage-free houses and are looking forward to receiving New Zealand Super for the rest of their days. Auckland graphic artist Toby Morris recently dealt with the squeeze his generation is under in Rent Rage.
What does this apparent difference in generational resources today imply for the sustainability of income and consumption in New Zealand over the next two or three decades?
I have to admit that these thoughts are not based on an accurate picture of the level of income, expenditure, wealth and fiscal costs of different generations and across time because we don’t have the data broken out by age. We know from the work of Bryan Perry at MSD that the over-65s at present suffer notably less poverty, but we can’t put any scale on this or understand it across time.
Over the past couple of decades, economists from around the world have used a standardised approach for digging into how people of all ages in different countries share resources. The approach focusses on the transfers between individuals, and between the government and individuals that arise from an imbalance between labour income and consumption in a particular year. This framework is called the National Transfer Accounts (NTAs) and in its simplest form it disaggregates the conventional National Accounts of Income and Expenditure by single years of age and calculates the various types of transfers, also by age. If you assume these transfer profiles don’t change much through time, then the values in one year can be regarded as the flows over the lifecycle.
The NTA framework is an international methodology created by Ronald Lee and Andrew Mason. The framework assumes that all our labour income equals consumption plus savings plus net transfers plus net income earned from assets. The identity holds at the national level as well as at the individual level. The difference between consumption and labour income at each year of age is called the lifecycle deficit. This deficit is balanced in each year for each year of age by transfers and capital income. From that starting point, researchers are taking a wider perspective: NTAs are being broken out by gender, wealth and time-use accounts are being added, as are accounts for different socioeconomic groups.
In fact, looking at the distributions of income and spending at each age may be more useful than just averages. For example, there may be a big intergenerational transfer from Baby Boomers who are "underconsuming" because they are not "reverse mortgaging" their freehold homes and therefore may leave considerable bequests, allowing their children to purchase homes, but at a much older age than previous generations. So the real problem may be that lifetime outcomes of members of future generations may become more determined than in the past by their parental outcomes. That's hidden when only comparing averages across generations.
In New Zealand’s case, we could use the data in the Household Economic Survey to create about 60 detailed age profiles of detailed spending (from ages 0 to 89, along with the group aged 90 and older), income, capital income and saving, which would add up to the lifecycle identity, as an analytical tool for intergenerational analysis. The unit of analysis is the individual and so modelling is required to allocate household income and spending to the individuals in the household.
At this point, New Zealand doesn’t have a complete set of NTAs, but some work has been done along these lines in the past. In 2004, for example, Jacques Poot and Sandra Baxendine used several Statistics New Zealand surveys (New Zealand Income Survey and the Household Economic Survey for 2001/02) to build a set of accounts for five-year age groups and the group aged 65 and over. After calculating the average lifecycle deficit for each age group and how transfers balance everything out in 2001/02, they used population projections out to 2051 to see how the aggregate results might change through time with high, medium and low scenarios for fertility, mortality and net migration.
Earlier this year, the Treasury invited a French economist (Julien Navaux) to look at our National Accounts and survey data to see how well we are equipped to construct NTAs for New Zealand. He was a member of the team building the NTAs for France.
Julien was very encouraged with what he saw in New Zealand. The major difficulty for constructing NTAs is the availability of data for individuals. Compared with many other countries around the world, he told us that New Zealand has a “precious asset” in the Household Economic Survey, which has been run from 1973 to the latest 2016/17, underway now. There exist about 36 HES waves (some of recent ones are just of incomes, between the triennial large surveys). These produce a picture of expenditures and incomes, both necessary to construct age profiles for the NTAs. This would produce a rich time series of transfer accounts for New Zealand and enable users to see how trends might unfold over the next three or four decades.
Such a dataset would provide an analytical platform for addressing many useful policy questions. Analysing a time series of NTAs, for example, would increase our understanding of how intergenerational transfers have changed historically, and the roles played by the public sector, families and the private sector in creating these changes. We could then use this work to project how profiles might change over the next half century. This would allow us to quantify changes in generational burdens given the current median population projections in, say, long-term fiscal or overlapping-generations modelling frameworks.
We could use the data to examine the extent to which private households offset public transfers through their own intergenerational transfers, bequests and changes in their savings behaviour. Another use might be to compare New Zealand’s position on the population ageing spectrum with that of the nearly 60 other countries with NTAs and see what policies countries with older populations than ours have adopted to address fiscal and economic pressures.
Research with this data should shed some light on what changes might be needed to ensure that Gen Ys will have sufficient income and assets to maintain growth of the New Zealand economy into their mid-lives.
NTAs would improve the quality of policy analysis and social policy debate. It’s easy to think of analytical topics around population ageing and its effects on the economy, the public accounts and on well-being generally that NTA data would help address. Moreover, we see the average age profiles as a public good, available to anyone for social and economic research. Building the detailed accounts is a resource-intensive exercise and is not a high priority in the Treasury as present. We should remain open, however, to the possibility as population ageing has a greater effect on incomes and spending by the private and public sectors.
The Treasury welcomes feedback on this article. If you intend to email comments or questions on this article please note these provisions.
Treasury Staff Insights: Rangitaki
See Treasury Staff Insights: Rangitaki for other articles in this series.