The changing face of ageing: From baby boom to baby bust

Editor's Notes

• #3 As background: The UK population has been growing at the rate of x % per cent annum By 2040 it will be 72m as compared with 48m in 1945 Not all natural increase includes considerable about so immigration but the effects will be largely baked into births and deaths The number age 65 will increase to 17.2m or 24% of the adult popn. Life expectancy has increased as well although it has stalled in recent years and affect by Covid. A negative feature is the wide inequalities but these not for discussion today
• #4 My starting point will be the baby boomers. There is a perception for example that there has only been one boom and that was immediately after 1945 but that would be incorrect. In fact the have been several at regular intervals not obviously of the same intervals or amplitude. These baby booms echo at regular intervals and can used for example to predict house building cycles because when women have their first baby at around 33 it is it is the time when they want to set up home or a year or two before. The second thing is that earlier baby booms helped predict the timing and number of deaths at points in the future although the links are complicated by mortality differences by age and improvements in mortality over time over time. As longevity has improved this has the effect of postponing deaths for some – age of deaths also used to be different between men and women but this is changing as the converge. Indeed if we ever get to a rectangularised survival curve and deaths become predictable all this could be easy but at the moment it is not. I therefore think that the historical pattern and timing of births and deaths can tell us a lot - I will direct my comments on two things: The nature of the cycles both births and deaths and what they tell us How the patterns we observe can inform pension age, house purchase decisions and inheritance
• #5 Lets me start with baby booms. In most peoples’ minds there was only one baby boom which came and went after the world war. In fact what this chart shows is that there have been several – I count four and predict a fifth between around 2041. I’ll spend the next couple of slides look at the pattern in more detail.
• #6  First I divide them up into trough-to-trough cycles rather than peak to peak. Your will see why in a minute. The table below the chart gives the details They are quite regular in occurrence for the specific reason is that successive booms are echoes of the previous boom as women start families – average age has increased but having first child is now around 30 in the UK. The peaks coincide very closely with other life cycle decisions such as house purchase, and planning cycles like maternity services child care and early years education. The peaks are broadly symmetrical about 25-26 years apart although the gap is increasing slightly probably reflecting that women are having children later The fifth cycle E which is an iteration of D which is informed by ONS population projections. My analysis suggests next trough will occur in 2030 and peak in 2041. Peak amplitudes abate over time but the number of births in each cycle pretty constant except for in B when it was much the highest
• #7 Before continuing there is one curiosity that is worth mentioning. As many of you know marketing organisations are keen on using names like baby boomers, Gen X and Gen Y to define specific cohorts. I superimpose their durations at the top of the chart. The concepts are widely used to explain the timing of later life events such as going to university or pension age and retirement. For example the convention is that baby boomers were born between 1946 to 1965 The problem is they have become very ingrained in public conversation but also that they are becoming increasingly implausible. I suspect they will breakdown as we get to Gen Z and of course after that we run out of letters of the alphabet! I think the idea originated in the US – for example boomers are 1946 to 1964. That definition works on a peak to peak basis for the US and also fits the UK quite well as the chart shows. Gen X on the other hand is defined as births between 1965 and 1980. That doesn’t work as well in the UK if your align it with the actual birth cycle. For the UK Gen X works better if defined as 1965 to 1977. Gen Y, the so-called millenials, are supposed to be people born between 1981 to 1995 in the marketing world. In the UK this charts shows that 1979 to 1991 works better for Gen Y. It probably sounds like I am splitting hairs but if you are policy maker or involved in producing age sensitive products accuracy is important since any mistakes are will be cumulative as the generations are supposed to butt on to each other and grow increasingly out of alignment.
• #8 Lets now introduce deaths into the story– roughly speaking there are around 600k a year in the UK give or take. These are shown by the grey series under births. These are probably more of interest to our audience today but the interaction with births is important. For most years except in1980 (a trough year between cycles B and C) deaths have been lower than births. An excess of births over deaths is usually a necessary condition for an increasing population not withstanding net migration. It has fluctuated over the years but most intriguingly hits a minimum in 2011. This is due to a combination of cohort effects and improved life expectancy. The first baby boomers reach 65 in that year. However, the number of deaths keeps climbing thereafter. As the first boomers age further more then die each year. We find that from 2033 the number of births and deaths are the same and from 2041 deaths surpass the number of births clocking in at about 800k a year –good news therefore for providers of long term care and funeral directors you might say but also for the secondary life market. Joking aside basically we can expect much to change that society will change pretty drastically as a result. The age of inheritance will be influenced in large part by life expectancy and the age of the mother when her child is born. A woman born in 1950 who has a child at age 20. That child would be 65 when they inherit in 2024. If the mother was 30 when the child was born then she would be only 55. Wind the clock forward 20 years, with increases in life expectancy those ages would be 67 ad 57 – two years older than if she had been born in 1950 so not by as much as many would have thought it seems.
• #9 Lets see what happens in a few specific cases: My first example is the dependency ratio which is used as an indicator of an ageing society and sufficiency of working age population to support and older population in pensions and health, and social care. Like most international agencies I use a reference age of 65 as the cut off. The dependency ratio in my case is defined as the number of people aged 20-64 to those aged 65+. Of course it can be inverted in which case the indicator becomes the percentage of older people relative to working age. The chart shows how it has varied down the years. There was a big change in 1980 at the end of the Gen X birth cycle after which it levelled out at 3.7 for the next 30 years or so. Once the first baby boomers had begun to reach 65 it started falling very quickly and will continue to do so at least until 2040. We have also had a lot of net in-migration especially between 2000 and now which has had the effect of making the DR more buoyant than it would otherwise be. Can it fall further? In a hypothetical rectangular population pyramid it could be asymptotic to one but this seems unlikely – levelling at around 2 looks more reasonable but it’s still a long way off.
• #10 Lets no see what that change will do to state pension age. As you know we have a multi-pillar system and everybody gets a PAYG state pension at age 66, the so-called ‘universal state pension age’. Currently it is 66 years and was 65 but this is reviewed every 5 years and next review is due now and so it could change again. This chart shows how the DR varies with SPA which is on the horizontal axis based on the latest demographic data. If you align it with the current value of the DR of 3.44 and you follow the hatched line across you get the current pension age of 66 Point P. Suppose the policy aim is to adjust future SPA so that the DR is constant over time. If you then wind the clock forward to 2030 updating the population in the process you find that the SPA needs to 68. If you further wind it forward to 2039 it needs to be 70. Present plans envisage SPA going up to 67 from 2026 and to 68 from 2037, but this charts shows that the increases need to come sooner than planned. This is bound to be the focus of the pending review and the outcome could be very contentious. Some of the more awkward options will include raising contributions, reducing benefits or asking people to work longer. So that’s pension age in a nutshell lets look briefly at housing markets.
• #11 Normally the bigger the population the more homes that are needed. However, there is no direct translation from the population size to the homes that are actually built. In fact it is hugely modified by the age distribution of the population and of course depends on whether it is family household with children or single person living alone. The blue line in this chart show the a average size of household a person lies in given their age based on UK data. As your can see it rises at first before declining with age. I call it the Dwelling index – which is something a developed a few years ago. As your can see a maximum occurs around age 10 with average of 4 persons – two parents and two children most likely. As you age further this goes down as first the children leave home and then possibly an adult dies. If they don’t move home and many do not the household becomes small indeed – tending towards one person living alone. However, there is a small uptick at the end of life as some people move into care homes. The chart superimposes a bedroom index which shows that average number of bedrooms in a household for people of varying age over the life course. Except under age 20 there is usually a bedroom surplus which occurs when the red line is above the blue. So in an ageing population what tends to happen is that average household size reduces so there is under utilised space i.e. surplus bedrooms. This is such as serious problem that I estimate that 11m will be living alone by 2040 up by 4m from 2000 with something like 22m spare bedrooms rattling around. Now all this is occurring during a period of acute housing shortages, massively inflated house prices and young people unable to afford to buy their own homes. I won’t go through possible solutions here but just to say this problem won’t end any time soon.
• #12 And this chart shows why an ageing population exacerbates shortages. There are three indexes beginning in 1980 with a stating value of 100 – population, household projections and occupancy. It shows that the number of households will increase faster than the population over time – shown as the two lines above the x-axis. Meanwhile average occupancy will decline meaning that house have to be provided or built to accommodate the same population. It is difficult to convey this in one chart, but one telling statistic from this is that if everyone lived in homes appropriate to their age 50k fewer homes would need to be built each year. If more retirement homes were built this would clearly help but it would need to be on a much bigger scale than now.
• #13 Let me try and summarise – I have shown how baby booms work through the population and the problems that arise in a few areas of public policy. I could have extended my examples to other sectors like health and long term care which are also age-related services. So these final remarks are meant by way of summary: There are patterns in the data which are not used or anticipated nearly enough – e.g. recurring baby booms of declining amplitude but equally spaced peaks of ~25-26 years Population ageing has an impact on state pension age which could rise to 70 before 2040 ahead of announced plans. In which case it is fair to say our government have got a grip The increase in deaths as baby boomers die out will trigger a tidal wave of inheritance which hopefully be used to the benefit of the following generations and should give lots of scope for financial services in particular. This is not being anticipated nearly enough especially by government in my view. Last but not least, despite the historic failure to build enough homes at in previous baby booms the ageing population will prolong the housing shortfall despite the release of many homes as the first baby boomers die away.