The geography of housing affordability

One of the important problems that we are facing these days is the high and rising cost of living. In particular, many households need to spend a substantial share of their monthly income on expenditures related to housing, either in the form of rent or mortgage. This problem affects many social groups, but probably at most the younger persons, who usually earn less and more frequently are unemployed, and those who live in larger urban centres. This is because apartment prices and rents tend to be much higher in urban than in rural areas.

The relationship between the cost of housing and household income is called housing affordability. For many reasons, it is crucially important to keep housing as affordable as possible. For example, high costs of housing in some cities or regions may discourage people from looking for more attractive job opportunities. Also, housing-related expenditures might affect long-term life decisions, like the decisions about having children. When housing costs consume a large part of the budget, other spending, for example on participation in culture and leisure time activities, will likely be reduced.

There is actually a range of factors that contribute to making housing less affordable. Housing markets, although local in their nature, tend to be dominated by few influential developers, who might form a kind of a cartel and dictate higher prices. Also, the nature of housing as a time- and capital-intensive good makes its supply inelastic, and in that way makes it difficult to adapt to changes in demand. And obviously, land speculation also plays a certain role, too. Given these circumstances, it is not surprising that many governments have undertaken attempts to make housing more affordable. However, it would be much beyond the scope of this short post to provide even a very selective review of such policies.

The specific combination of socio-economic factors and policies leads to different levels of housing affordability in different countries, regions, and cities. This important fact, although seemingly well known, is in my opinion only to limited extent explored empirically, and perhaps due to that reason it tends to be overlooked in discussions. In my opinion we need to better know the differences between particular countries, regions and cities in terms of housing affordability, and try to understand the reasons behind these differences. In this post I would like to give (at least a partial) answer to the first part of this question. The latter part might be in my opinion an interesting subject of further discussion.

In this post I would like to present a brief comparison of housing affordability among a group of countries. The first question I had to address was: which source of data and which indicators should I choose? Among the available options, I chose the “OECD Better Life Index” database. It is particularly useful in this context because it contains an indicator that measures the percentage share of income that an average household needs to spend on housing. In that way, it is a very good measure of housing affordability. However, it might happen, for example, that households in a particular country have low housing expenditures, but at the same time they live in small apartments. Therefore, from the same source I took another indicator that measures the average number of rooms per person. Both indicators were plotted on a chart, which is shown below. In order to make the analysis more systematic, some arbitrary criteria were introduced that allow us to specify which countries are characterized by the best (or worst) affordability of housing.

(click on chart to enlarge)

Therefore, in the group of countries with the best affordability I included these where not more than 20% of income is spent on housing, and the average number of rooms per person equals at least two. The following countries fall into this category: Australia, Belgium, Ireland, Norway and the USA. As we see, it is a very differentiated group of countries, representing different parts of the world. They also have strongly varied population densities, ranging just three persons per square kilometre in Australia and around 30 in the USA, to over 350 in Belgium. This suggests that that the scarcity of land, in contrast to what seems to be widely believed, does not necessarily determine the affordability of housing. Notably, another high-density country – the Netherlands – is very close to the criteria of “best affordability” specified above.

On the other side we find countries that are characterized by the worst affordability of housing in the selected group. In that case the criteria of selection were the following: housing expenditures amount to at least 22% of income, and the average number of rooms per person does not exceed one and a half. Five European countries meet these criteria, among them three post-socialist states (Czech Republic, Poland and Slovakia), and two countries from the Mediterranean zone (Greece and Italy). The composition of this group is probably a result of several reasons, and some country-specific factors likely play a significant role. However, one common factor seems to be a relative weakness of institutionalized housing policy. In particular, the post-socialist states have tended to adopt a neoliberal approach, in which the main responsibility for housing supply was shifted to the private sector, and the state has to a large extent taken a passive role. On the other hand, one needs to remark that among the post-socialist states Slovenia and Estonia perform somewhat better.

The weakness of state housing policy appears also to affect the situation in the case of Greece and Italy, which traditionally tended to rely more on a family-based welfare model. However, in the group of Mediterranean countries there is also a remarkable divide, as particularly Spain is characterized by a large number of rooms per person and a moderate housing cost. That might be interpreted as a result of the housing boom, which greatly increased the supply of new apartments, and the following bust, which led to a decline of prices. Another interesting case is Portugal, where the number of rooms is comparable to that of Germany or France, but the cost of housing in relation to income is one of the lowest among OECD countries.

Finally, we can shortly refer to some curiosities. A very specific case is Russian Federation, which, as in the exemplary case given above, is characterized by extremely low housing expenditures (just 11% of income), but simultaneously has a very low number of rooms per person (only 0.9, which is the lowest value among the all included countries). Among the post-socialist states, Russia seems to be most representative of the old centrally planned model, in which housing was provided at low prices, but also in small quantities (and usually at low quality). Also, a remarkable case is Brazil, which is situated close to the middle point of the chart, next to countries like Austria, France, Germany, Iceland and Sweden. In that case, it seems questionable to what extent the official data reflect the reality of a country that is known for having a high share of population living in favelas.

To conclude, this very brief overview obviously cannot cover the complex issue of the regional differentiation of housing affordability in a systematic way. My purpose was rather to provide an introduction, and possibly to encourage a discussion. In my opinion this issue appears to be of substantial importance. Although the “death of geography” was (prematurely) announced some years ago, we see that in today’s world differences between places do not seem to disappear. Instead, they may even tend to increase. And particularly housing affordability will play a crucial role as it affects both the competitiveness of countries, regions and cities, and the well-being of their residents. So, we should look more intensively into this topic, also at the level of regions and cities.