The state of the housing market in the euro area

Prepared by Niccolò Battistini, Julien Le Roux, Moreno Roma and John Vourdas

Published as part of the ECB Economic Bulletin, Issue 7/2018.

The housing market has important macroeconomic and macroprudential implications for the euro area economy. In view of the duration of the ongoing upturn in euro area house prices and residential investment, which started at the end of 2013, analysing the state of the housing market is particularly informative. This article discusses the ongoing housing market upturn, from a chronological and fundamental perspective. It also explores a selected set of indicators that can potentially inform on the state of the housing market, elaborating on the demand and supply factors underpinning the current upturn, as well as their relative importance.

1 Introduction

Understanding and monitoring the state of the housing market is important because of its macroeconomic and macroprudential implications. Housing market developments affect investment and consumption decisions and can thus be a major determinant of the broader business cycle. They also have wealth and collateral effects and can thus play a key role in shaping the broader financial cycle. The housing market’s pivotal role in the business and financial cycles makes it a regular subject of monitoring and assessment for monetary policy and financial stability considerations.^[1] This is especially relevant given that housing markets can be the source of booms and busts, with severe and long-lasting consequences for economic and financial development. Such episodes tend to reflect a decoupling of expectations over housing market tendencies from their fundamental determinants.

The housing market has a price and a volume dimension. Residential property prices (hereafter “house prices”) and residential investment are relevant dimensions and are the main focus of this article. They can be seen, in a broader context, as outcomes determined by the interaction of different supply and demand factors. Price and volume developments are not necessarily synchronised, so that possible misalignments between them can be an additional source of information. However, they can also make the overall assessment of the state of the housing market more challenging. For the euro area, this assessment is subject to the caveat that there is considerable heterogeneity across housing markets and their developments across countries. In addition, disentangling developments that could be associated with past boom/bust episodes, with a period of accommodative monetary policy or with changes in structural factors adds to the complexity of the analysis.

The state of the housing market is, by nature, unobservable but can be assessed from different perspectives. From a chronological perspective, the state of the market can be characterised by the length of its upturns or downturns, in particular, in comparison with the average durations of such phases. From a fundamental perspective, it can be assessed by the position of key indicators relative to benchmarks, for instance to determine possible house price overvaluation or unsustainably high activity in construction. The set of indicators that can potentially inform on the state of the housing market is large and the discussion in this article therefore needs to be selective.^[2] It aims to distinguish broadly between demand and supply factors, although even this distinction may be difficult given the nature of some of the indicators.

Against this background, this article has two main sections. Section 2 puts into perspective the recent developments in house prices and residential investment in relation to the business cycle. Unless otherwise stated, these two indicators are expressed in nominal and real terms respectively. Section 3 elaborates on the demand and supply factors underpinning the current upturn in the housing market, as well as their relative importance.

2 The state of the euro area housing sector: a look at residential investment and house prices

The upturn in the euro area housing market is in its fourth year. Measured in terms of annual growth rates, house prices started to pick up at the end of 2013, while the pick-up in residential investment started somewhat later, at the end of 2014. The latest available data (first quarter of 2018) indicate annual growth rates above their long-term averages (see Chart 1, left-hand panel) for both indicators. This is more evident for residential investment (where the upturn in growth rates has levelled off) than for house prices (where it has continued). The timing of the start of the upturn is broadly the same when measured in terms of the levels of the two indicators. At the same time, the level perspective highlights that residential investment is still considerably below earlier peaks, while house prices have recovered from the declines recorded during the financial crisis (see Chart 1, right-hand panel). In the aftermath of the financial crisis, residential investment declined sharply by 25%, bottoming out in 2014. Thus far it has only partially recovered and in early 2018 was still 15% below its pre-crisis level. House prices, on the other hand, contracted by only 6% between peak and trough, and in early 2018 were standing 5% above their pre-crisis level (although in real terms – deflated by the HICP – they were 5% below their pre-crisis level).

The upturn in the housing market is common to the majority of euro area countries. Over the last four years almost all of the countries have witnessed positive average growth in both residential investment and house prices, although with different magnitudes (see Chart 2). On balance, the majority of the countries share the feature, observed for the euro area as a whole, of concurrent growth in investment and house prices during the current upturn (blue dots), mirroring the relative adjustments in the preceding downturn (yellow dots). Some natural questions arise. How prolonged is the current housing market upturn compared with historical regularities? And, what can we expect going forward?

A turning point analysis suggests that the housing market upturn is in a relatively advanced phase compared with the average duration of such upturns. From a chronological perspective, at the aggregate euro area level, the length of the current upturn in both residential investment and house prices of about 4 years amounts to the average duration of historical upturns in residential investment and half the duration of historical upturns in house prices (see Table 1). For the purpose of this article, an upturn exceeding half the average duration of previous upturns shall be considered relatively mature. It should be borne in mind that turning point analysis is surrounded by considerable uncertainty, especially at the end of the sample, and that there is no unique way of dating the economic phases. Notwithstanding these caveats, if we consider the five largest euro area countries, the residential investment cycle has reached a mature phase in Germany and the Netherlands, while it is still at an early stage in France, Italy and, to a lesser extent, Spain. According to this metric, the house price cycle is likewise advanced in Germany, while it is still at a rather preliminary stage in the other countries. In Italy, the formal turning point analysis does not yet suggest an upturn in house prices.^[3] The maturity of the upturn can also be related to the strength of the recovery, with more mature cycles generally exhibiting lower rates of growth compared with those recorded at an early stage of the cycle.

The current euro area upturn is stronger than historical averages for residential investment but weaker for house prices. If cycles were to evolve around an unchanged trend, the relatively strong upturn in residential investment can be related to the relatively large fall in the aftermath of the crisis: during the ongoing upturn, euro area residential investment has increased at an annual average rate of 3.5%, clearly above the average of 1.9% recorded for the same duration in previous upturn phases (see Table 1). For euro area house prices, the corresponding comparison suggests a relatively muted upturn, with an annual average rate of increase of 2.7% – below the historical average of 6.3%. For house prices, this relatively muted pattern is common across the largest euro area countries, whereas in the case of residential investment the outcomes are mixed: the Netherlands and France exhibited higher than average upturns, while the opposite was true for Germany, Spain and Italy. This metric is an additional gauge for assessing the state of the housing market, but it comes with the caveat that the relative strength of the upturn may look “artificially” low in countries where the historical averages are influenced by unsustainable booms in the housing market. To this end, assessments against fundamental values are also needed.

Assessing house prices and residential investment against fundamentals also provides insights on the state of the cycle. Chronologies of the housing cycle can only provide a partial gauge for assessing the state of the housing market, although, empirically, being out of tune with fundamentals often coincided with the state of the cycle being relatively more advanced and having seen a relatively strong magnitude of adjustment. Benchmarking against fundamentals can take several forms, such as simple ratios, deviations from model-explained values, or synthetic combinations of such metrics. In the case of house prices, valuation estimates are regularly applied in financial stability assessments^[4] and currently point to a slight overvaluation for the euro area as a whole, as well as for Germany and France (see Chart 3). In the case of residential investment, this article introduces a synthetic indicator constructed from different (standardised) ratios for output and employment in the construction sector. For both the euro area and the largest euro area countries, this indicator suggests that the residential investment cycle is close to its historical norm, somewhat above for Germany and somewhat below in the case of Italy.

Overall, the state of the euro area housing market is relatively mature but is not, so far, characterised by disproportionate residential investment activity or house price dynamics. The analysis shown in this section suggests more strength and maturity in the euro area residential investment cycle than in the house price cycle. However, measured against underlying fundamentals, the former does not appear, so far, to be above its historical norm. Given that the housing sector can be an important driver of the business cycle and that residential investment and house prices can have leading indicator properties for future economic activity, the current state does not herald imminent risks of a move towards a contraction in the economic cycle (see Box 1 for a more detailed analysis).

3 Supply and demand factors behind the current state of the housing market

House prices and residential investment can be seen, in a broader context, as outcomes determined by the interaction of supply and demand factors. Such underlying factors can thus shed additional light on the state of the housing market. However, corresponding indicators are scarce, often lagging, and are not always easy to interpret in terms of whether they provide information unequivocally on the demand or the supply side. This section makes a selective approach to discussing some of these indicators.

3.1 Demand factors

Consumer survey indicators point to ongoing increases in demand for housing. During the current upturn of the housing cycle, the number of respondents intending to carry out home improvements and to purchase or build a home has gradually increased in the euro area and in the vast majority of euro area countries (see Chart 4). The latest data thus suggest that further demand for housing – related to both the stock and the flow (investment) – may still be in the pipeline. For the euro area as a whole, the intention to carry out home improvements was close to an all-time high in mid‑2018, while the intention to purchase or build a home had increased more modestly and remained well below pre-crisis peaks. Since housing-related sentiment indicators have improved, on balance, more moderately than overall consumer confidence, data do not seem to point to a risk of exuberant demand. Intentions to purchase or build a home reflect a combination of cyclical and structural factors. Box 2 discusses homeownership as an example of the latter.

The rising demand for housing has been supported by developments in income and financing conditions. The current euro area housing upturn has been accompanied by an expansion in real disposable income. In addition, financing conditions remained favourable, as reflected in composite bank lending rates for house purchase that have declined by more than 130 basis points since 2013 and by easing credit standards. This has given rise to a higher demand for loans for house purchase and a substantial strengthening in new mortgage lending (see Chart 5). The expansion in loans for house purchases net of repayments has been rather moderate and thus suggests that the upturn in the housing market came with more moderate increases in mortgage indebtedness. However, gross loan origination suggests, at the same time, that the actual availability of credit for the purpose of purchasing and building houses is more than ample.^[14] In this respect, the growth in mortgage loan origination in the euro area has been more synchronised with the growth in house prices. From a cross-country perspective, loan origination is currently at historical highs in Germany and France, close to its historical average in Italy, while it remains subdued in Spain.

Housing demand is likely to have been supported also by investment motives. The relative attractiveness of housing as an investment class has increased during the recent housing upturn. Estimates of the return on housing-related investment are surrounded by considerable uncertainty but suggest an increase in the relative attractiveness of investment in residential property vis-à-vis alternative asset classes – such as government bonds, deposits and equities – since 2013 (see Chart 6).^[15] Private and institutional investors, both domestically and globally based, searching for yield may thus have contributed to additional housing demand.^[16] One expression of this search is portfolio reallocation and flows into real estate funds, which have increased steadily for the euro area as a whole since the beginning of 2013, also as a share of residential investment (see Chart 7). While too small in terms of size to account for substantial shifts in overall demand for real estate properties (some of which could also be directed to commercial real estate or outside the euro area jurisdictions), these funds can nevertheless indicate additional housing demand for investment purposes.

From a longer-term perspective, the positive cyclical factors of housing demand may have been dampened by structural factors such as demographics. Since 1995, the decreasing growth rate of the 20‑49 population age group, which is an important cohort in terms of housing demand, can contribute to explaining the declining residential investment share of GDP in the euro area and may have exacerbated the sharp cyclical fall in residential investment following the onset of the crisis (see Chart 8). Looking ahead, projections of growth in the euro area 20‑49 year-old population bracket suggest a bottoming out of this dampening structural factor in the coming years, leaving more room for ongoing positive cyclical forces to fuel demand for residential investment (as a share of GDP). Over a longer period of time, the relationship between the growth rate of the 20‑49 population age and residential investment is also observed at the country level.^[17]

3.2 Supply factors

This section analyses how the volume of houses has evolved in recent years and the extent to which there are factors which constrain housing supply.

Housing completions in the euro area have remained substantially below their average level since the start of monetary union. This indicator can be viewed as a measure of the flow of new houses supplied to the market. In Germany and France this flow has recently been close to the average levels observed since the start of monetary union, whereas in Spain, Italy and, to a lesser extent, the Netherlands it has remained subdued (see Chart 9). At the same time, the granting of building permits, which are a necessary – but not sufficient – condition to building a house, has increased more strongly than housing completions in a number of large euro area countries. Since it appears that the supply constraints from a lack of building permits have been easing, increases in residential investment activity and new housing supply may be forthcoming.

The lack of building permits can be a constraint to housing supply. In central locations and cities, especially, the number of issued permits may fall short of actual demand due to the scarcity of building land. A relatively low number of permits then pose a constraint on the supply of new homes. In central locations, where the availability of land is limited, the competition for scarce building permits may be aggravated by demand from investors interested in building commercial real estate, and who – according to anecdotal evidence – currently seem to have a preference for urban areas. However, data limitations prevent firm conclusions from being drawn.^[20] The notion of local supply constraints is supported by a stronger increase in house prices in capital cities vis-à-vis the corresponding countries’ average in the current housing market upturn.^[21]

In the shorter term, housing supply can also be constrained by the time required to receive permits. Administrative restrictions – such as the time required to obtain a building permit – are a major factor affecting the elasticity of housing supply in reaction to demand.^[22] Consequently, when demand for housing picks up, one would expect it to show initially in a relatively larger rise in house prices than in quantities, as measured by residential investment. To illustrate this point, in 2017 a construction company needed roughly 126 days to obtain a building permit in Germany, compared with 228 days in Italy and 208 days in Spain (see Chart 10).

Another factor that may have limited housing supply is a shortage of labour in construction production. Survey data on the percentage of construction firms signalling constraints on production due to a lack of workers suggest that labour shortages have started to become an issue in the current housing market upturn. The percentage of companies reporting labour as a factor limiting production recorded a fourfold increase in the euro area over the past three years, from almost 5% in the third quarter of 2015 to more than 20% in the third quarter of 2018 (see Chart 11). Among the five largest euro area economies, these developments were more accentuated in Germany, France and the Netherlands.

Labour shortages may affect the prices in the housing market in different ways. On the one hand, shortages may lead to gradually higher wages in the construction sector, which, in a situation of sufficient demand, will then be passed through to construction output prices. On the other hand, shortages may constrain or delay the supply of new houses in relation to demand and then imply a rise in house prices that does not necessarily come with higher construction output prices. Thus far, the marked rise in construction costs since 2014 has been increasingly fuelled by rising labour costs (despite their lower weight in the overall index), indicating growing labour shortages, and has been accompanied by strong momentum in house price growth, signalling buoyant demand (see Chart 12).

3.3 The relative importance of housing demand and supply

Assessing the relative importance of housing demand and supply factors is intrinsically challenging, as data mainly refer to equilibrium outcomes. For example, rising house prices may reflect an increase in demand for housing or a reduced supply of houses. With this caveat in mind, this subsection reviews available survey evidence and model-based results to assess the relative contribution of supply and demand factors to the state of euro area housing markets.

Survey data point to an increasing relative importance of supply factors in determining the dynamics of prices and investment in the housing market. The proportion of construction firms^[23] indicating constraints to their production has hovered between 50% and 60% since 2014. However, the composition between reported demand-side and supply-side constraints has varied significantly: the share of firms reporting insufficient demand as a factor limiting production has decreased substantially, with opposite developments for firms reporting supply-side constraints (see Chart 13). Recently, there has been approximately a 10‑percentage point increase in the number of respondents noting supply-side rather than demand-side constraints limiting production. This suggests that constraints on construction producers’ output have recently mainly come from the supply side. This evidence is confirmed through the lenses of a stylised model with residential investment and house prices (see Box 3).

The relative importance of housing demand and supply factors can also be assessed by means of information concerning the composition of house prices. The more demand outpaces supply, the more prices of existing dwellings should be expected to rise as the competition for housing extends to existing properties. Whether the prices of existing dwellings rise faster than those of new dwellings also naturally depends on whether there are bottlenecks in the supply of new dwellings and on the responsiveness of construction output prices. It is thus conceivable that the prices of existing dwellings reach a larger amplitude at the peaks of the cycle compared with the prices of new dwellings. Indeed, over recent years, the contribution of prices of existing dwellings to overall house price growth has risen sharply, from close to 10% at the start of the upturn in 2014 to almost 80% in 2016, hovering above 70% – but on a declining path – over the last year (see Chart 14). At the same time, the increase in the contribution of prices of new dwellings since late 2016 (although still subdued) may confirm a tightening housing market, with increasingly binding supply-side constraints: amid buoyant housing demand, new dwellings cannot be provided fast enough and their prices tend to rise more rapidly.

The analysis presented in this section suggests an increasing role for supply-side constraints in determining dynamics in the euro area housing market. During the early stage of the upturn, a significant positive adjustment in residential investment was accompanied by relatively smaller house price increases. Over the most recent quarters, as the upturn in the housing market continues, buoyant demand amid increasing supply constraints has been associated with moderating residential investment growth and continued rising growth in house prices.

4 Conclusions

The euro area housing market has been in an upturn since the end of 2013 and is in a relatively advanced state of the cycle in terms of duration. House prices have surpassed their pre-crisis peaks, while residential investment is still significantly below. The state of the euro area housing market is, so far, not characterised by generalised investment activity or house price levels above their fundamentals. However, considerable heterogeneity in developments across and within countries makes the overall assessment more challenging.

The housing market upturn is expected to continue but at a more moderate pace. This reflects expectations in currently available forecasts and projections that the euro area economic expansion will continue, reflecting the favourable impact of the very accommodative stance of monetary policy, improving labour market conditions and stronger balance sheets. This context generates income and financing conditions conducive to housing demand. Lending to households for house purchase is also expected to remain dynamic in the coming years. Nevertheless, in line with the expected slowdown in the pace of economic activity, the rate of expansion in the housing market is also expected to moderate. A moderation in residential investment might also emerge from the increasing presence of supply-side constraints in some euro area countries, which may currently be more binding than in the respective economies as a whole. These constraints could however mitigate the envisaged moderation in house prices.

Monitoring a broad set of housing-related indicators is key to assessing the macroeconomic and macroprudential implications of the housing market. To fully assess the state of the housing market it is necessary to look at both the major demand and supply determinants and their interactions. Moreover, given the extended interactions between real and financial variables, a broader set of indicators – some of which were discussed in this article – that goes beyond house prices and residential investment (such as loan developments, house price valuation, household balance sheets, etc.) should be continuously monitored to fully understand the macroeconomic and macroprudential implications of the ongoing housing upturn.