Notes One

EXPLAINING GROWTH DIVERGENCES IN THE EURO AREA: THE ROLE OF RESIDENTIAL INVESTMENT
Klaus - Jürgen Gern and Carsten - Patrick Meier
IfW, Kiel Institute for the World Economy


In our econometric analysis of housing investment in the euro area we, as a first step, distinguish between Germany and the rest of the euro area. This distinction is motivated by the fact that German developments have been special in several respects over the past 20 years and, given the large size of the Germany as a proportion of the euro area economy, influence the developments on an aggregate euro area level significantly (as opposed to, e.g., Austria where housing investment—and population growth—was rather similar to German patterns).

Since swings in housing investment in Germany over the recent two decades have been associated with pronounced changes in population growth, special emphasis is given to the importance of demographic developments in explaining diverging trends in housing investment. Population growth and demographic developments that affect the number of households, such as shifts in the share of population in household formation age and falls in the average size of households, all have potentially important implications for housing demand.

We estimate functions for Germany and the rest of the euro
area where residential investment depends on the existing level of housing
stock, real income, the user cost of capital (proxied by the real interest rate) and
population growth.

In the literature, the impact of demographic developments on the housing market is usually discussed in the context of house prices.1 Several studies include demographic variables in their estimated house price equations (e.g. Meen 2002, Ahearne et al. 2005) , but they are omitted in recent VAR-studies (Sutton 2002, Tsatsaronis and Zhu 2004). Girouard and Blöndal (2001) discuss the influence of house prices on residential investment through the price-cost ratio, but do not discuss other determinants of residential investment.


Residential investment is typically a relatively volatile demand component, characterized by pronounced swings in growth rates from year to year and subject to strong cyclical fluctuations.

It is evident that, notwithstanding a large amount of heterogeneity, national
series tended to move in similar directions. This is especially true for recessionary phases (in terms of aggregate output), such as the mid 1970s, the early 1980s and early 1990s, although the timing of the peaks and bottoms varies slightly from one country to another.

In the most recent phase of weak production growth between 2001 and 2004, there is also some similarity in that most national series show at least some deceleration of growth. However, the strong deceleration of residential investment growth registered for the euro area as a whole is to a large extent due to a prolonged and pronounced decline in residential investment in Germany; housing investment in other large euro area economies such as France, Italy and Spain by contrast remained relatively strong.

While in Germany investment stagnated in the second half of the 1990 and has been falling since 2000, Spain, the Netherlands and France saw a rapid expansion of investment; in Italy investment declined in the second half of the 1990s but picked up after the start of EMU.

The standard deviation of growth rates of residential investment in the euro area is much larger than the standard deviation of aggregate output growth, reflecting the relatively pronounced volatility of this demand component. Although there is no clear trend discernible in the standard deviation over the past 40 years, it is interesting to note that differences in growth rates have been relatively small in the most recent years, while the divergence of economic growth in general and residential construction activity in particular in the years following the German unification is clearly visible.

The post-unification boom in residential investment in Germany has been associated with a sharp acceleration of population growth stemming mainly from immigration from Central and Eastern Europe and refugees from the Balkans.3 Conversely, when residential investment decelerated in the midnineties and started to decline towards the end of the century, population growth slowed to a crawl. Inspection of the relationship between population growth and residential investment growth (smoothed using centered 5-yearaverages) over a longer period of time and across other euero area countries shows, however, that such a close co-movement of these variables is not the rule.

In 1990-91, German unification led to a de-synchronization of cyclical developments, triggering a boom in Germany when most other industrial countries experienced deceleration of growth or even recession. Although some of the demand generated in Germany spilled over into neighbor countries, economic growth in the rest of (what is today) the euro area slowed down when German growth accelerated. Following a brief period of almost synchronized growth in 1993 (recession) and 1994 (recovery), from the mid-1990s onwards the German economy consistently grew less rapidly than the rest of the euro area.