2.1. Ageing and house prices

Before surveying recent work on ageing and house prices, we briefly introduce considerations relevant to modelling of house prices.

The main approach to house price modelling has been to rely on hedonic price models with these being developed since the 1960s. Rosen (Reference Rosen1974), building on Becker (Reference Becker1965) and Lancaster (Reference Lancaster1966), developed a framework that allowed the prices of goods to capture different attributes which generate utility for the consumer. This approach allows for product differentiation. Unlike homogenous goods, homebuyers are matching their preferences to a product with a unique set of characteristics—an important consideration for house purchases since when purchasing a property, consumers consider the utility benefits of a host of characteristics, such as location, dwelling size, neighbourhood characteristics and so forth. However, these multiple factors will also impact the price of a property because they possess monetary value. As Chin and Chau (Reference Chin and Chau2003) note, homebuyers are purchasing a product that is a set of attributes and changes in one of these factors will change the utility benefits and thus the marginal price. Since the total house price is the sum of all marginal prices, the change in an attribute of one house relative to another explains differences in house prices.

The advantage of the hedonic approach is that it does not treat houses as homogeneous goods and it allows the researcher to add attributes to a baseline model, including demographic considerations that are now important in the context of ageing populations. Thus, demographic factors can be modelled in terms of their impact on house prices and implications for other economic variables can be inferred.

However, a major problem of the Rosen approach is that the price equation is non-linear and no specific functional form is provided. This means prices derived from the equation can vary depending on the chosen specification. Typically, in a first stage, prices are regressed on the commodity attributes to obtain estimated marginal prices for each. These are then used in a second-stage regression to estimate an inverse demand function, assuming the marginal prices are endogenous to demand. Moreover, it is assumed that housing supply is unaffected by consumer demand, which is unlikely to be the case (Chin and Chau, Reference Chin and Chau2003).

Hedonic models can also pose additional problems when evaluating the impacts of ageing demographics on house prices. They cannot easily account for unobservable neighbourhood characteristics, such as standard of local care provision, transport accessibility and disability access. These are important considerations for aged homebuyers, so their exclusion could lead to biased inferences. Elderly homeowners may spend significant sums on property renovations but lack of data on this means their impact on house prices could create omitted variable bias. This can also create inaccuracies in house price estimation since hedonic models often assume depreciation can be proxied by the age of a property but older properties that have been renovated by the elderly will depreciate less or may even appreciate in value.

Another approach, utilised in several studies cited in this paper, is the asset model of house prices. As equity prices are related to the discounted value of future dividends (dividend discount model), so house prices should be related to discounted net rental values. The intuition behind the Poterba (Reference Poterba1984) ‘user cost’ approach is that owning a property (via purchase) should be financially equivalent to renting. If renting is cheaper, demand for ownership falls as do house prices. If purchase is cheaper, rental demand falls. Hence under a no arbitrage equilibrium, the price of a property should equate to the present value of discounted rents on equivalent properties. Rents then become dependent on house prices which are in turn dependent on factors such as real interest rates, depreciation, property taxes and expected house price inflation.

In Poterba et al. (Reference Poterba, Weil and Shiller1991), the above approach is used to specifically incorporate ageing demographics by splitting cohorts into subsets by age with the elderly being classed as over 65 years. The average effect of more elderly populations is a reduction in long term house prices since the more aged are assumed to exit the market or not renovate. However, in the short run, because housing supply is relatively inelastic and because expected deflation of house prices is slow to adjust, prices may not decline significantly.

Since property is the major durable asset holding of many households, they provide the biggest utility gains in an average household’s asset portfolio. However, unlike homogenous equities where equity risk premia are commonly estimated, in the case of housing risk premia is less understood. Other problems with the user cost approach to determining the impact of ageing arise from omitted variable bias. As previously mentioned, data on renovations are sparse. Any impact of ageing may be capturing other excluded factors such as credit constraints in the lending market and macroprudential restrictions like loan-to-value limits.

As described hereafter, authors such as Arestis and Gonzalez-Martine (Reference Arestis and Gonzalez-Martine2017) assume short-run supply of housing as fixed and augment the Poterba (Reference Poterba1984) model by including macroeconomic factors such as unemployment and real disposable income alongside demographic variables. Other studies, as discussed below, have incorporated elderly homeownership, elderly wealth accumulation and decumulation.

Studies of effects of ageing on house prices can be divided between those using calibrated overlapping generations or life-cycle models and those using econometric techniques. The former are generally based on calibrated behavioural coefficients, typically applied to the behaviour of individuals inter alia in their demand for housing, and in some cases also firms and government. Meanwhile the latter are based on statistical estimation, in the field of housing demand generally at a global, national or regional level.

In more detail, overlapping generations and life-cycle models typically examine shifts in the assumed financial behaviour of individuals identified inter alia by age, aggregated across a wider population, often within in the context of a small economy model. They generate average or economy wide results of ageing on financial behaviour including housing demand. Effects on housing demand in the context of ageing affecting prices are typically only one of the outputs of these models, as seen in the examples, below. Life-cycle models differ from overlapping generation models mainly in a focus on an individual’s lifetime saving and consumption choices while the latter analyse interactions between generations also. On the other hand, econometric studies typically focus on the determination of house prices or related variables related to housing demand such as home ownership directly, including ageing as well as other determinants of global, economy wide or regional level house prices or home ownership. These studies are assessed in greater detail in Davis and Karim (Reference Davis and Karim2025). We focus on each in turn.

As regards studies using overlapping generations, Cerny et al. (Reference Cerny, Miles and Schmidt2010) proposed a calibrated overlapping generations model including ageing, housing and pension reform. People decide in their model whether to buy or rent property and borrow against it, as well as receiving labour income (or in retirement, a state pension), and deciding whether to hold safe or risky financial assets. Results suggest growth in financial assets over the working life followed by reductions in retirement, but with some retained for bequests. This is preceded by a rise in debt that is decumulated over the working life. Owner occupation peaks around 60 years of age but does not decline greatly till late in retirement, which as discussed below seems to be borne out by some econometric studies. Owner occupation and house prices both increase in future years, but the stock of financial assets rises by more.

Lisack et al. (Reference Lisack, Sajedi and Thwaites2021) prepared an overlapping generations model of the advanced countries that includes house prices and household debt as an output as well as real interest rates. It was suggested that longevity drives increases in wealth, as individuals seek to provide for a longer retirement period. This in turn reduces the real interest rate both in the past and the future (assuming longevity trends continue) and raises the capital–labour ratio. Lower interest rates and longevity account for a large proportion of the increase in real house prices up to 2008, and also the doubling of household debt/GDP ratios as young people borrowed to buy more expensive houses.

Yogo (Reference Yogo2016) used a life-cycle model of the US, examining housing choice in the context of health depreciation and potentially costly healthcare. This may be especially applicable to European countries if social care in old age is expensive. Retired consumers select levels of consumption and, health expenditure, while undertaking asset allocation between bonds, stocks and housing. Retirement income arises from social security and defined benefit pension funds. The model shows that the portfolio share in housing is negatively related to declining health and declines as people get older, in line with time horizons. A pension reform (a one-time reduction in social security benefits) leads to a shift from housing and equity to bonds. Health expenditure is seen as a form of investment in health capital but with decreasing returns; out-of-pocket health expenses as a share of income are higher for those in poor health and older.

Turning to studies using econometric methods, an early study by Mankiw and Weil (Reference Mankiw and Weil1989) of US housing demand created much interest, as it suggested a major fall in house prices by 47 % between 1990 and 2010 as baby boomers retired. The result was not borne out as house prices continued to rise. Age and cohort effects had not been sufficiently distinguished, as baby boomers sought more housing services than earlier generations, so did not ‘trade down’ as predicted.

More recent studies have taken this critique into account via identification of cohorts or the use of pooled or panel data. For example, Igan and Loungini (Reference Igan and Loungani2012) in a model of real house price changes for a large set of countries used pooled estimation from 1970 to 2010. They included working-age population along with other typical house price determinants and found the change in the working age population was significant and positive for house prices.

Davis et al. (Reference Davis, Fic and Karim2011) modelled the demographic effect in more detail by using those aged 20–39 as a share of population (the main house buying cohort) for a panel of advanced countries. Its size was significant in boosting house prices, although only for the UK, US and Canada.

Takáts (Reference Takáts2012) used data on 22 developed nations from 1970 to 2009, using panel estimation. He found that the change in the elderly dependency ratio (negatively) and the change in population (positively) had significant effects on real house price growth. Annual increases over the sample from changing demographic shifts were estimated to be 0.3% per annum over the historic sample. He suggests that population ageing in coming decades may reduce real housing values by about 0.8% per annum but this is likely to be offset by economic growth. Gevorgyan (Reference Gevorgyan2019) also found population growth boosts house prices for 59 countries up to 2018 but did not test for other demographic variables.

Jäger and Schmidt (Reference Jäger and Schmidt2017) estimated house price determination over 1950–2012 in 13 developed countries. They found a negative effect of the over-65 cohort and a positive effect of pre-65 s, but no relation of house prices to life expectancy or population size. It was suggested that demographic patterns depress house prices both in the past and the future, but in the past, this has been compensated by GDP growth; ageing plus subdued growth might lead to more actual house price falls in the future.

As in Davis et al. (Reference Davis, Fic and Karim2011), Arestis and Gonzalez-Martine (Reference Arestis and Gonzalez-Martine2017) looked at house price determinants in 17 OECD countries over 1970–2013, using age groups 25–44, 45–64 and 65+. Effects were assessed country-by-country. Results showed different demographic effects between countries, with, for example, the US having long-run and short-run positive effects from the two working age groups (25–44 and 45–64) but no effects from the retiring group (65+). Australia had short- and long -run negative effects from the retired group. No demographic effects were detected for the UK. Accordingly, cross country panel studies may hide some important national differences in housing market dynamics.

One of the few studies that look at the effect of ageing at a local as opposed to a national level and also at submarkets within the overall housing stock is Wang and Durst (Reference Wang and Durst2023). They examine the impact of ageing as proxied by the elderly dependency ratio at a US county level. They find that there is an overall boost to house prices from a larger percentage of elderly people, but that this is focused in particular on smaller homes (one bedroom) and there is no significant effect on condominiums and larger houses. They suggest results at a national level may be distorted by differences in ageing and also zoning rules at a local level and that the pattern found relates to downsizing.

On balance, the results of estimates for prices of non-financial assets (e.g. property) are that a larger elderly cohort restrains housing demand. But in most cases, the literature does not suggest an outright decline. It appears that house prices are more likely to respond to specific features of demographics in the countries concerned than equities, since the latter are much more internationally traded. The studies may underestimate the role of housing supply, which if limited, could raise upward pressures on house prices in future.

2.2. Ageing and home ownership

As regards shifts in the relative demand between real and financial assets with ageing, it seems that housing demand would fall less than for financial assets as the elderly cohort grows. This is due to risk preferences, and the dual need for housing as a producer of consumer services and as an asset, alongside the illiquidity of housing. This is consistent with the results for house prices cited above which suggest the size of the elderly cohort has relatively little impact on prices when other factors are taken into account.

Such a view is underlined by research of Chiuri and Japelli (Reference Chiuri and Jappelli2010) who looked at elderly homeownership across 15 OECD countries. Whereas the raw data suggest a decline in homeownership from age 60 onwards, controlling for cohort effects mitigates this decline considerably, instead it begins at age 70 and accelerates at age 75. They find that mortgage market regulation (including scope to extract equity) is a key explanation of differences across countries—in fact it has a negative correlation in that those able to extract equity are more likely to remain homeowners than those who have limited scope to do this.

Of course, equity extraction implies a decline in net wealth while remaining as a homeowner, whereas ceasing to be a homeowner implies a switch to the rented sector, retirement or care home. As noted by Bogataj et al. (Reference Bogataj, McDonnell and Bogataj2016), however, equity release schemes are often assigned to the property and not the property owner and hence may limit desirable downsizing. Property taxes and availability of care beds were not significant determinants of home ownership by the elderly.

Meanwhile, Horioka and Ventura (Reference Horioka and Ventura2024) investigated homeownership’s role in a wider assessment of the pattern of wealth accumulation/decumulation by the elderly. Whereas the life-cycle hypothesis assumes that the elderly will decumulate wealth to maintain consumption in retirement, the evidence shows that less than half of old people in European countries are decumulating wealth, with the average rate of accumulation being 6.6% over a three-year period. The explanations include a bequest motive (keeping wealth for the next generation), generous public pensions (that reduce the need to decumulate) and reluctance of the elderly to sell their houses or borrow against their accumulated equity.

The bequest motive has a greater effect on accumulation, the larger the expected bequest. Precautionary saving in Europe seems to be less important than in the US, which may relate to public health systems and lesser concern about longevity risk due to public pensions. Homeowners are more likely to be accumulating wealth than renters, thus underlining that reluctance of elderly homeowners to sell homes or borrow against equity helps to explain the decumulation puzzle. It is suggested that consumption of the elderly can be increased via higher inheritance taxes, as would availability of home equity loans.

Similarly, Blundell et al. (Reference Blundell, Rawford, French and Tetlow2016) find a lack of decumulation in the UK even by those aged 70–74; housing tends to be retained in retirement unlike in the US (partly due to house price increases) although financial assets are also decumulated less rapidly than in the US.

3.1. The reluctance of elderly to downsize

The results of the empirical studies outlined above clearly have certain policy implications. If ageing will lead to declines in property prices, then there is a risk of negative equity for mortgage borrowers, which in turn could impact on lending institutions. On the other hand, the balance of the literature on ageing and house prices seems to suggest that declines in house prices due to ageing will be limited when other factors are considered. This is underlined by the work cited on home ownership and asset decumulation by the elderly which suggest a very limited rate of decline of homeownership as people age, especially when cohort effects are allowed for. This occurs in a wider context of a strong bequest motive (in Europe) and precautionary saving (in the US and Japan), which entails that financial as well as housing assets are decumulated slowly if at all, while the elderly accumulate equity in their houses as prices rise.

On the other hand, there are important corollaries of these results which are less discussed in the empirical literature. For example, the tendency of elderly people to continue living in large homes (e.g., 76% of over-65 s in the UK are owner occupiers) is a key issue for the distribution of housing across the population. A question raised in the UK is whether the reluctance of the elderly to leave their existing homes is related to the lack of suitable homes to move to, be they bungalows in the open market or specialised sheltered accommodation for older people, with or without care. Note in this context that specialised retirement housing is only 2.6% of the housing stock while 90% of the elderly live in open market mainstream housing. This issue is particularly pertinent since 84% of projected household growth between 2018 and 2043 involves people over 65.

In this context, Mayhew (Reference Mayhew2020) estimates thatFootnote ¹ there are more than 15 million unused bedrooms in the UK, of which 60% are owned by the over-65 s. Under occupation is accordingly concentrated among elderly singles and couples, but there is a lack of age-friendly housing in the UK even for those willing to consider downsizing; LGA (2022) suggest that 3 million people are ready to downsize in the UK but are put off by lack of choice. Only 7000 retirement homes are typically built each year, which is dwarfed by the expected rise in over 65 households to 2030 of 180,000.

Even in the US, where downsizing is more common, Wang and Durst (Reference Wang and Durst2023) note a shortage of affordable elderly housing that warrants public intervention. Also, they note the demand for smaller homes as a consequence of ageing may conflict with similar demand for small homes by first time buyers.

Absent policy action, Mayhew (Reference Mayhew2020) suggests that in the UK, the ‘bedroom surplus’ is expected to rise to 20 million in 2040 with 13 million of the over-65 s living in unsuitable accommodation. This is based on existing estimates (as noted above) projected forwards using UK demographic forecasts. It is suggested that getting the elderly into appropriate housing will not only free up space for first-time buyers but also reduce the burden on the health service and social care as social care can be more efficiently deployed in sheltered housing and retirement communities. Stays in hospital can accordingly be reduced.

3.2. Barriers to construction of retirement housing

Why do builders not construct more retirement homes, notably in the UK? The answer must address barriers to building houses generally, as well as specific problems giving rise to barriers to building retirement homes. We note that a number of these issues arise more widely than the UK, as noted by the World Economic Forum (Masterson, Reference Masterson2022), although some details are specific to the UK.

Factors underlying slow housebuilding in general include planning delays (Badii et al., Reference Badii, Brinkmann, Datta and Kulka2024). These in turn link to factors such as the need for sizeable financial contributions to the local area (to finance roads, other local infrastructure and services to cope with the increased demand from extra housing), infrastructure delays, cost of meeting regulatory requirements (such as environmental protection), delays to council decisions and community opposition. These have been particular issues for large developments that in recent years have accounted for a larger percentage of applications than in the past.

Meanwhile, costs for housebuilders have increased sharply in recent years, arising from land, labour, materials and financial costs. Skilled labour is in short supply. Resultant pressure on profit margins reduces incentives to build, particularly in the case of affordable housing which has lower margins (Masterson, Reference Masterson2022). Economic uncertainty arising from Brexit and COVID-19 have also served as disincentives to housebuilding.

There is also an argument that developers prefer margins to volume more generally, resulting in a slow release of houses to the market to keep prices high, so-called drip feeding (Roberts, Reference Roberts2025). A particular problem arises for affordable housing as local authorities are limited in their ability to build more, and much of the stock of social housing has been sold off under ‘right to buy’ legislation.

As regards the supply of retirement housing, Stirling and Burgess (Reference Stirling and Burgess2021) cite barriers, such as a lack of suitable sites, difficulty competing with other land uses due to higher costs than for general needs housing (such as wheelchair accessibility requirements), as well as planning barriers such as those arising from inconsistencies between local authorities around planning use classes for retirement homes. They contend that there is a lack of strategic vision to older people’s housing needs at a national level and poor assessment of needs for retirement housing at a local level. There are also constraints to investment in retirement communities such as long investment time horizons, which may deter interest by quoted investment funds. Meanwhile, housing market stimulus packages are typically focused on incentives and housing supply for young first-time buyers rather than the incentives to supply retirement homes and for elderly homeowners to downsize.

Mayhew (Reference Mayhew2020) also cites a major change in 1990 when the Community Care Act required local authorities to help vulnerable adults to stay in the community, and which led to a rationalisation of the retirement housing sector and a collapse in supply, especially by local authorities and not-for profit organisations, to at most 25% of its peak ever since.

Demand factors also play a role in discouraging supply. We have already cited the reluctance of older people to move from their existing homes. They are likely to be attached to a local community whereas retirement communities are often remotely situated and with poor access to public transport. And they do not necessarily wish to move to smaller homes. For a couple where one person needs social care, care home fees in the UK are levied on financial assets of that person and not on the value of the property, which creates a further disincentive to downsizing.

Leasehold contracts offered to retirement homes are much less attractive than freeholds people hold and may dissuade people from selling their existing homes. It is not clear that the recent reform to replace leasehold with commonhold and limit ground rents will make retirement homes, or flats more generally, more attractive (Hudson, Reference Hudson2026).

There are often exit fees to sales of retirement properties embedded in contracts, which may be 10%–30% of value in some cases (Age UK, 2025). A further key aspect is maintenance fees associated with retirement flats as opposed to houses that discourage movement. Crawford (Reference Crawford2026) cite a large number of people unable to sell the retirement homes they have inherited, and who thus pay very large such maintenance fees. Such aspects justify individuals with concerns about inheritance to desist from purchasing retirement homes. Taxes, especially stamp duty land tax, also militate against moving house to downsize to retirement housing, as discussed below.

Outside London and the Southeast, Stirling and Burgess (Reference Stirling and Burgess2021) suggest that ‘retirement communities’ appear to be unviable; this is constrained to areas where older movers can access more housing equity, which limits choices for those living elsewhere. Indeed, they suggest that only social renters and wealthy homeowners seem able to access retirement housing.

In light of issues as identified above, Mayhew (Reference Mayhew2020) suggests that local authorities should require housebuilders to include suitable housing for the elderly in planning applications, as they do for affordable housing. More efficient use of the existing stock, with the elderly living in suitable housing as opposed to remaining in, and underoccupying, their existing properties, would reduce pressures to ‘just build more’. Bogataj et al. (Reference Bogataj, McDonnell and Bogataj2016) further highlight the need for local demographic projections to play a greater role in planning for housing, notably for the elderly.

Further policy action suggested by Stirling and Burgess (Reference Stirling and Burgess2021) might include aligning accessibility standards of mainstream and specialised housing, and support for specialised housing applications from health and social care authorities (as it promotes efficiency in delivery of their services). Given individual preferences, housebuilders may need to consider a wider range of options for elderly individuals seeking retirement housing than one-bedroom retirement homes and guarantee long leases. A strategic vision for elderly housing is needed at a central and local government level.

3.3. Tax systems, housing and ageing

The tax system in many countries also militates against downsizing. For example, in the UK, stamp duty land tax (a transactions tax) on house purchase at rates of 5%–12% on properties valued above £250,000 disincentivises downsizing. As noted by Whitehead et al. (Reference Whitehead, Crook and Scanlon2024), this tax means people seeking to downsize cannot buy a property of the same value as their existing one without paying the tax, whereas staying in their existing home avoids such costs. They argue that elderly people downsizing should have the same incentives as first-time buyers, with greater exemption from stamp duty, or its deferral till death.

Progressive ad valorem property taxes regularly revalued and thus properly indexed to inflation would provide a much better incentive to downsize, if they could replace transactions taxes. Bogataj et al. (Reference Bogataj, McDonnell and Bogataj2016) identify similar issues in Spain (and more widely in EU countries). The UK is a particularly egregious example of unindexed property taxes, as property valuations have not been revised since 1991. As noted by Muellbauer (Reference Muellbauer2025) the poorest pay the highest tax as a proportion of their income and higher band rates are much higher in poor areas such as Nottingham than rich ones such as Westminster. This, combined with high rates of stamp duty land tax, implies in turn that elderly people in high value properties in richer areas have little incentive to move. However, the wholesale transition to a revalued and indexed system may be difficult, given it will create winners and losers.

In what would be an incremental but also radical reform, Muellbauer (Reference Muellbauer2025) argues that the two top rates of council tax in the UK should be replaced by an annual wealth tax of 0.5%, which would be 1% for non-UK taxpayers and second homeowners. There would also be a 1% tax on agricultural and unoccupied land valued above £40,000 a hectare. This would affect an estimated 1.1 million properties, thus limiting political risk. There would certainly be an incentive for pensioners in these top-rated properties to downsize if this proposal were to be introduced, even though they would be allowed to delay payment of the tax till the property is sold. A fall in high value property prices would then likely ensue. Muellbauer estimates that the proposal would raise £9bn in revenue from the top two council tax bands, which could be balanced by a reduction in stamp duty land tax to 5%.

Whitehead et al. (Reference Whitehead, Crook and Scanlon2024) recommend an overall reform of the council tax system, which would entail revaluation of all properties to current values, and adjustment of the bands in a way to imply that those with higher-valued properties pay proportionately more tax. The authors suggest that such a reform would increase the tax burden on those making the largest untaxed capital gains and increase the incentive to move to lower-valued properties by downsizing.

Adam (Reference Adam2025) of the Institute for Fiscal Studies similarly recommends reforms to council tax to be proportionate to current valuations. It is suggested that rates should be set to replace revenues from both current council tax and the stamp duty land tax and that both of these should be abolished. Such a reform could again encourage downsizing. Adam (ibid) also notes that the current 25% discount from council tax available to sole occupants encourages individuals to remain in large houses and recommends either its abolition or setting not be in line with property values.

Meanwhile, without reforming the underlying system of council taxation, the UK government has raised a surcharge on properties valued at over £2 m from 2028, to rise in line with inflation. 200,000 properties may be affected, raising around £400 m. This could lead to more decumulation of property by the elderly (although as in the Muellbauer proposal, they are to be given the option of deferring liabilities till the property is sold).

3.4. Issues in rental for the elderly

We may add that the taxation of owner occupation is also generally much more favourable than for landlords with homes to rent, with the latter typically subject to income tax on rental income and capital gains tax. This offers a further disincentive for homeowners to sell up and rent, since the disparity in taxation is reflected in the cost of occupation.

Some perspective on these policy issues can be provided by noting that despite the current generation of elderly being unique in terms of its prevalence of owner occupation, more and more elderly people are renting. In the UK, for example, older renters are predicted to double in the next 20 years (Age UK, 2024). 30% of older renters live in ‘non-decent’ homes, over 50% of renters approaching retirement (45–64) have no savings and may find paying rent in retirement difficult. And 70% of over-65 renters would find it difficult to find an alternative property if they had to move.

Australia has seen similar patterns, with overall owner occupation falling from 71% to 67% in the past two decades. Amongst the elderly, the 75+ group has the highest owner occupation and the 60–64 group has the highest proportion of renters (Chen et al., Reference Chen, Tran, Sinclair and Boymal2024). Renting has increased by 73% among the elderly in the past decade.

Such patterns raise the issue of whether the overall system of retirement tends to assume mortgage-free owner occupation (with just maintenance and property taxes to pay) and what the implications of more renting will be as ageing proceeds. Certainly, it may raise the required level of income needed for ‘comfortable’ retirement.

3.5. Ageing and financial regulation

Turning to financial institutions and financial regulation, Mitchell (Reference Mitchell2018) lists several desirable innovations to allow for the effects of ageing linked to housing. These are based on US experience but are clearly relevant elsewhere. Notably, to allow for wealth shocks, it may be helpful to have reverse mortgages; risk sharing in defined contribution funds between workers and retirees; and more inflation-indexed assets. Development of financial literacy will be crucial, as will allowance for declines in cognitive abilities with age.

In this overall context, the question arises whether housing can be made more liquid for the elderly to aid their financial sustainability in retirement? What are the obstacles to trading down, equity release and so forth and are they amenable to regulatory action? Does it point to a need for re-examining regulation of equity release or reverse mortgages? In light of the analysis of ageing and housing, does regulation of mortgage lending warrant review, in case of issues such as age-discrimination in mortgages (which is forbidden in the US) and issues in trading down/equity release (as noted above)? It is notable that in Australia, the government offers a home equity access scheme to supplement pension income which can be obtained as a lump sum or a monthly income stream, at a rate of 3.5% (Chen et al., Reference Chen, Tran, Sinclair and Boymal2024). The US have the HECM programme, which guarantees reverse mortgages and is highly used (Bogataj et al., Reference Bogataj, McDonnell and Bogataj2016).

Meanwhile, in the context of affordability for older people currently confined to renting, Chen et al. (Reference Chen, Tran, Sinclair and Boymal2024) suggest fractional ownership models in housing could be a helpful way of owning real estate, facilitated through blockchain. It allows multiple buyers to share ownership of a property and could be a viable option for older Australians who want to access the benefits of homeownership without the high upfront costs and risks of actual 100% purchase.

Relevant research on the UK mortgage market by Bogiatzis et al (Bogiatzis-Gibbons et al., Reference Bogiatzis-Gibbons, Charles, Lawlor, Limani, O’Brien and Staniland2025) from the Financial Conduct Authority found no evidence of discrimination of mortgage pricing by demographic characteristics (including age). However, they could not conclude that there are not issues with ‘demand fairness’ via the availability of products to different groups. There are other potential drivers of differential outcomes, such as the impact on the supply of mortgages of the financial conditions of individuals in certain social groups at the point of taking out a mortgage.

3.6. Interaction of housing and pensions

Interaction of housing and pensions is a further key issue as ageing proceeds. Building from their model cited above, Cerny et al. (Reference Cerny, Miles and Schmidt2010)) suggest that a reduction in real pensions paradoxically boosts the housing market further, as people have more income in working life, due to the assumption that the government will levy lower taxes on labour income to pay public pensions. Later retirement also boosts housing demand as people have more lifetime income, whereas saving for (and during) retirement is lower.

An analysis of the issues for retirement-income risk raised by the rise of defined contribution pensions in the UK is given by Boileau et al. (Reference Boileau, Cribb and Emmerson2025a, Reference Boileau, Cribb and Emmerson2025b). Investors have to manage the investment of a lump sum received on retirement. This issue comes particularly to the fore since, as discussed previously, the option of downsizing or equity release from owner occupied housing is unpopular and little used. A model of ‘flex then fix’ with drawdown up to a certain age and then annuitisation of the remainder is seen as a possible way forward as a non-compulsory default option.