Spanish housing review by RICS
The housing market downturn continued during 2009. There was some evidence that the decline might be bottoming out towards the end of the year, though the housing market and the wider economy still face substantial problems. So, it is by no means certain that the end of the housing market downturn has actually been reached.
The adverse market conditions have been feeding into official price indices. There is uncertainty and debate about the extent of the price falls that have occurred, especially in new build. As a result, it is worth spending a little more time than usual in the other country chapters to go through the different indices.
There are two official indices. The most long-lasting is the Housing Ministry index based on the property registry, which means that it significantly lags market transactions. It recorded an 8% year-on-year fall in q3 2009 (figure 10.1) and a 10% decline since the 2007 price peak (and a 1% fall on the previous quarter). There are also some recent indices from the national statistics institute, INE, based on data from notaries.
INE indices on new and existing house prices indicate divergent outcomes. The results from the new-build housing index are surprising and may reflect issues with the nature of the collected data, because the price series shows virtually no change since the end of the boom, which seems at variance with the continued existence of very large stocks of unsold new dwellings (with more in the pipeline), popular commentary and the heavy discounting currently being undertaken by the banks and major developers holding stocks of completed housing. (However, the efficacy of discounts as a measure of actual house price change is questionable, as they are marketing devices that tend not to be based on current market values.) As a result, only the existing homes series is shown in Figure 10.1.
By contrast, the INE series for existing homes shows marked annual house price falls of 11% in q2 2009 and a decline from 2007 peak of 16% (Figure 10.1). Other price indices also highlight significant price falls. But they are not that out of line with most of the official indices. Fotocasa’s house price index recorded a 10% price fall in the year to November 2009, and a 20% decline from the peak. Tinsa’s index, based on valuations, showed the most modest decline at 6.6% year-on-year to November 2009, with an overall 11% fall from the peak.
Overall, an approximation of a 15% price fall from the peak of the boom would seem to be a reasonable interpretation of the prevailing house price data. Taking account of general price inflation, this suggests a roughly 20% real decline in values. There are undoubtedly substantial locational and dwelling type variations around this average national number. This will be due to differences in both local speeds of adjustment and the amount of house price change generated by revised relationships between local demand and supply.
The country had one of the biggest housing market booms in Europe up to 2007. Property prices increased by 2.2 times in real terms between 1996 and 2006; housebuilding rose to record heights; and mortgage debt increased dramatically. Housing investment alone was 8% of GDP in 2006 and construction as a whole, much of it related to real estate, about 13% of GDP. Average household debt reached 130% of personal disposable income in 2007 – three-quarters of it related to mortgages – and in 2009 was still 125%, although falling interest rates had lowered the current burden of outgoings.
The extent of the market slowdown can be seen clearly in transactions data. The number of registered sales had peaked in 2006, some time before the onset of the financial crisis. But by the third quarter 2009, recorded sales were 40% less than that peak (Figure 10.2). Though it is hard to measure the vacancy overhang, overall unsold and soon to be completed dwellings are believed to amount to over a million units; a proportion of which are holiday homes but many are primary residences.
Housebuilding tells a similar story of decline but from extraordinary peaks in the final years of the boom. Almost 600,000 dwellings were completed in 2006, which works out to be around 13 dwellings per 1,000 Spanish people – many times higher than the volumes most other European countries were building in their housing market boom years.
Since then, output has been sliding. However, completions did not start to fall until late summer 2008 and then surprisingly levelled out in 2009 (Figure 10.3). This may be due to statistical issues related to when a completion is recorded, but also shows the time it takes to build out developments and the incentive for even heavily loss-making developers to keep on building if they can raise sufficient short-term loan finance and sales revenue to put off the threatened day of failure.
Housing starts (issued permits) reacted more quickly to changed market conditions. They had fallen to around only a fifth of their previous peak level by 2009 but, like completions, they flattened out in 2009 as well. Even at that reduced level, they were still significantly higher on a per capita basis than the building troughs of other stricken European countries, such as the UK. It may be reasonable to assume in consequence that the flow of new supply still probably outpaces much reduced housing demand.
There is an issue with the housing output data, in that completions remain far higher than starts to a degree that building time lags are unlikely to explain adequately. The data suggest that over 360,000 dwellings per annum are being completed over two years after a major turnaround in the housing market. Typically, the expectation in a severe housing market crisis is that developers would mothball sites with long development trajectories, either voluntarily or involuntarily through pressures from creditors. It may be that case that many developers, or their financiers in cases of insolvency, are being slow to declare completions due to the huge number of vacant dwellings already on the market. Whatever the true cause of the discrepancy, the supply-side does not yet seem to have fully adjusted to changed market conditions.
A continuing threat to the housing market is the overhang of unsold property that is currently being kept off the market by the behaviour of financial institutions. The cause of this recalcitrance may lie in the financial system, because banks have incentives to be slow in foreclosing on problem real estate development loans and to hold on to completed properties they have on their books acquired as collateral for defaulting loans. High loan volumes were granted to the real estate development sector in the boom years (to finance land, developments, underway, and finished housing). Although the market has obviously declined since then, the current true valuations of those heterogeneous loans and the underlying real estate assets can be argued to be subject to uncertainty, since there are no organised secondary markets for them. Accordingly, the practice has been adopted by lenders of a progressive recognition of impairment in which the actual loss in relation to amortised cost is posted in accounts only gradually over time as further evidence clarifies true values. Of course, banks can only do this if the loans and the real assets backing them are held rather than sold, because then the values would become certain, and the practice is kept within the limits of acceptability to the supervisory authorities, which have been quite accommodating in this respect.
The outcome of such balance sheet beneficial practices is that many unsold properties have been kept off the market by banks and developers, which has helped to moderate price falls. This experience has been distinctive from many other countries; as, for example, with the rapid foreclosure processes associated with defaulting owner occupiers and developers in US markets that triggered so much of their downward trajectory. It has helped to avoid a similar reinforcing downward market spiral in Spain, but sustained recovery is unlikely to occur until many currently unsold properties are absorbed by the market. In any case, as house price falls become clear and the fragility of developers’ balance sheets become increasingly obvious, the process of gradual write-downs begins to find its limits. A consequence may be quite a substantial flood of properties onto the market in the near future. Even if that flow can be regulated, such sales are likely to continue to put downward pressure on house prices for some time to come, with potentially adverse effects on price expectations.
The government has introduced some fiscal measures to revive the economy and offset the decline in the housing market. For example, a temporary measure has been introduced to enable the unemployed to roll-up interest payments rather than default and public expenditure on construction activity has been increased. However, the scale of the downswing is such that any actions are only likely to moderate decline. Some basic long-term drivers of housing demand are still strong. Acute housing shortages continue, despite all the building, because of previous low housing standards, demographic pressures and long-term rises in living standards. There have been high levels of immigration, raising the population by over 4 million in recent years, which has put further pressure on housing. But such factors in themselves count for little in an economic depression.
Housing markets are invariably heterogeneous locationally and by dwelling type and Spain’s diverse geography highlights such differences strongly. Consequently, the scale of the price changes that have occurred and the risks of the further downward adjustments vary across the country and by market segment. Nationally, some commentators argue that prices still have a long way to fall. For example, the leading consultancy Aguirre Newman argues that prices have to fall by nearly 30% more to get back to long-term affordability ratios3. Currently, low interest rates have improved affordability but they will increase in line with developments in the euro zone economy as a whole, rather than Spain alone, and unemployment is rising. House price forecasting on the basis of mean reversion to simple ratios was criticised in Chapter 1. Even so, clear risks remain with regard to the current price levels, despite nascent signs of housing market recovery.