Average house prices in the UK have risen much faster than average incomes over recent decades. Relative to average disposable incomes, houses are not far off three times as expensive now as they were in the early 1980s; relative to median incomes, they have risen even more.
This is not unique to the UK and it is far from uniform across the UK. In London and the South East, the rise has been more marked than in the rest of the UK and the scale of the rise in house prices there relative to incomes has been exceptional – though it is certainly not unique, and the trajectory of prices in San Francisco, Paris, Hong Kong, and New York is comparable.
UK house prices in a longer-term context
It is useful to set recent UK experience in a longer-term context. Evidence on house prices across a large sample of developed economies back to the mid-19th century reveals two things. Knoll et al. (2017) show that, on average, house prices moved roughly in line with the price of other things between about 1870 and the mid-20th century in most countries. Figure 1 (from their American Economic Review paper) shows the time series of cross-country mean and median real house prices for their full sample. Having shown no clear upward trend for almost a century up to around 1950, real average house prices then roughly triple over the next 60 years.
So there is nothing unique about the fact that in the UK house prices have in recent decades risen very much faster than the price of most other things and that houses have become dramatically much more expensive than cars, entertainment, food, clothes, electrical goods, energy, and travel.
Figure 1 Average real house prices across 14 developed countries
Source: Knoll et al. (2017).
Why has this happened? It is certainly not because the cost of constructing houses has driven up prices; that can account for only a tiny part of the huge rise in the relative price of houses. Increases in the price of land on which houses can be built, in contrast, have risen enormously. But then the question is why have land values risen so much.
Part of that answer is connected to panning restrictions – and some people are satisfied to leave that as the answer. But those restrictions are not arbitrary – they are neither random nor exogenous. Such restrictions exist in most countries. They are typically most binding in parts of countries where population densities and house prices have become high and they reflect underlying forces that make densities and valuations high.
The fundamental forces at work
In recent work with James Sefton, I have tried to assess the extent to which these more fundamental forces might account for the evolution of house prices and throw light on where they might go next (Miles and Sefton 2017).
Our focus is on the long run. We consider the housing sector within a model of the overall economy that aims to explain overall economic growth, saving, and asset prices. Any long-run analysis has to model the changing supply of housing taking into account the fixity of land mass and the way in which shifts in the cost of land relative to structures changes the way in which houses are constructed and where they are built. Land is obviously not homogenous and the impact of the most important way in which it differs (that is by location) varies over time as technological change means that distance may have a varying effect on value. One obvious way in which this happens is if transport costs change.
This rise in the relative price of housing across most developed countries in the period since WWII has come as the proportion of the population living in big cities has risen in most developed countries. It has also coincided with a period where real transport costs have been flat or (more recently) often rising; that is markedly different from the period between the middle of the 19th century and WWII when transport costs fell dramatically. These phenomena – rising relative price of housing, an end to falls in transport costs, greater urbanisation in population – are plausibly linked. Agglomeration forces – creating incentives to work close to other people – may have got stronger over the past 100 years; transport improvements have allowed people to live further from where they work. The changing interplay between those factors is a major force behind national and regional house price changes.
To understand that link, we use a framework that combines features of a Ramsey two-sector growth model with a model of the evolving geography of residential development that tracks the change in location of the population over time. We find that there is great sensitivity over time in the pattern of development, the types of houses built and the values of structures to even small changes in two key parameters: the degree of substitutability of land and structure in creating houses; and the degree to which households substitute between housing and non-housing goods.
We find that so long as improvements in travel technology proceed at a pace that is in a fixed proportion (of one half) to the growth in productive potential (the sum of labour force growth and general productivity growth), there is a balanced growth path with no change in real house prices. That condition was roughly satisfied in many developed economies between the mid-19th and mid-20th centuries.
But once travel improvements fall back – as they have done in more recent decades – we are no longer on a balanced growth path and real house prices and rents rise. How fast they then rise becomes dependent on a range of parameters that play little role when travel improvements were at a rate near half GDP growth. We then find that plausible parameter estimates plugged into this growth model can easily generate ever rising housing costs – relative to the price of other goods and to incomes. But there is great sensitivity of that to parameters that reflect both preferences (between different characteristics of houses) and technology. One key technology factor is how one combines structures and land to create housing. That has changed – the New York skyline shows that it is now possible to erect super-tall residential buildings on small plots of land and squeeze more residential space from a plot than was possible in the past. Whether that can fundamentally change the likely future cost of housing is an interesting question that is as much to do with consumer preferences as with construction technology.
Price sensitivity of demand for housing – reflecting the substitutability of housing for other goods and services in creating satisfaction – is another factor with a very powerful effect on the longer-term path of housing costs. Transport improvements have been very important in the past evolution of prices – improvements in commuting speeds have held down rises in the cost of housing. But we find that it is not likely that further improvements will be as powerful a force in keeping housing costs down.
We find that it is not difficult to find sets of assumptions about elasticities which are plausible in the light of existing evidence and which imply that house prices can rise faster than incomes for periods spanning generations. This has nothing to do with planning restrictions.
Ultimately, housing over the long term could become increasingly expensive and perhaps increasingly rented. That would be more likely if:
- population and productivity both grow steadily;
- people are unwilling to substitute away significantly from expensive housing towards other goods as housing costs rise;
- people do not substitute much away from using land in favour of space saving homes (high in the sky or underground) as land prices rise;
- there is no improvement in transport infrastructure and travel times. That would limit how far from the most densely populated and popular urban centres people can live.
These four conditions probably all hold in the UK. That sounds like bad news. But major investment in transport infrastructure that reduce commuting times could mean future house price rises can still be limited.