"Economists set themselves too easy, too useless a task if in tempestuous seasons they can only tell us that when the storm is long past, the ocean is flat again"
~ John Maynard Keynes
'As far as I am concerned, this is ... the most complex crisis we've ever seen due to the number of factors in play'
~ Spanish Economy Minister Pedro Solbes speaking this week to Spanish radio station Punto Radio
Jose Luis Borges tells a story about two rascally villains, eternal rivals, who - under sentence of death - are offered one last bet: rather than accepting a conventional execution they can agree to have their throats slit simultaneously, just to see who is a able to run the farthest. Immortality, rather than fame, in an instant. Now I mention this since tale I can readily anticipate the immediate feelings many will have on reading what follows (I am at the end of the day going to argue that it is necessary to inject money - and I do mean rather a lot of money - into a banking and construction system which many will want to argue is largely responsible for Spain's present distress, and indeed, that having made a good deal of money out of the operation, these are the very people who should now be forced to don that sackcloth and ashes costume which so behoves them (actually the way things stand they are much more likely to find themselves reduced to a sporting a loincloth, but still). I understand why many ordinary Spanish people may have such feeling, but I do think this is a time for cool heads, and that what is most needed here is an extreme dose of pragmatism coupled with a lot of emotional intelligence. There is no point in agreeing to have your own throat slit just to see people you don't like have their's slit first.
Martinsa Fadesa The First To Go
This week's filing by the Spanish property developer Martinsa-Fadesa for protection from its creditors has brought Spain's ongoing economic agony back to the headlines. The decision follows a request from Martinsa Fadesa last Friday to its creditor banks for a postponement of the deadline on their requirement that the company obtain a 150 million euro ($235.7 million) loan. The banks refused the request and the rest is now, as they say, history. The failure of Martinsa - Fadesa whose debts are in the region of 5 billion euros - is not only the largest corporate bankruptcy in Spanish history, it is also a reflection of the pain which must now be being felt in Spain's troubled banking and construction sectors, and a harbinger of what is, in all probability, going to be much worse to come.
Spain At Risk
So to come directly to the matter which has provided me with the header to this post, just what is the risk that the present recession in Spain is something a bit more than a mere recession? What is the risk of a real and serious economic melt down just across France's Southern border, a mere stone's throw away (by plane) from Brussels or Frankfurt, yet still on the other side of that intellectual and cultural divide which seems to be formed by that ever so picturesque natural barrier known as the Pyrennees? Well it is a non-negligable one, in my view. Let me explain a bit.
First, as background it would be worth reading my Artemio Cruz Syndrome post, since all the main macroeconomic arguments are presented there (and those who seriously want to know what is going on should definitely read the excellent "Spain:Bubble Bursting - We now expect a full-blown recession" desknote from PNB Paribas).
Secondly, we need a bit of vocabulary clarification, since the terminology being used has become somewhat confusing of late. We could reasonably break things down as follows I think:
i) Soft Landing
ii) Hard Landing
iii) Melt Down
Now, in terms of the available semantic space, why don't we allow that "soft landing" means a recession of the more or less garden variety (as Portugal or Italy have at this moment, or as say France may anticipate, or Denmark) and not consider this to mean avoiding recession completely, which is how some seem to have used the term in recent times (I think it is hard to imagine any EU 15 economy avoiding recession completely between now and Q2 2009). Possibly Hungary up to this point could also be said to have had a comparatively soft landing.
"Hard Landing", on the other hand would be what they are currently experiencing over in the Baltics, what they may well soon experience in Romania, Bulgaria, the UK, and Ireland, and what is now most certainly taking place in Spain. Thus by "hard landing" I mean a very sharp slowdown in growth, a medium sized contraction in consumption, financial distress and bankruptcy in some areas, and a recession which drags itself on for more than a mere two quarters (in and out of negative growth) and probably results in annualised negative growth for a period of at least 12 consecutive months. What happened in Turkey in 2000 was certainly a hard landing in this sense.
iii) "Melt Down", following such definitions, would then be a Hard Landing plus, a Hard Landing plus a shock (or in Hungary's case, where the shock would be a run on the forint, you could imagine what initially is only a Soft Landing being converted into a melt down, but arguably Hungary's case is very special given the very high level of exposure of household balance sheets to CHF denominated forex loans).
Such a shock could be a banking crisis, a run on the currency, a sovereign default (this is where Italy's series of perpetual soft landings could move decisively into meltdown mode one of these fine days if something isn't done to correct the low growth/high sovereign debt to GDP dynamic while there is still time).
Now in this sense, Spain's economy is at some significant level in danger of having a melt down - lets define this as more than two years of negative GDP growth with a magnitude of more than one percentage point, coupled with (in the case of countries which have their own currency) very sharp devaluations, and in the case of those that don't severe and extended price deflation (ie a mini version of what happened in the USA in 1930).
Now the recession in Spain is, I think, more or less most certainly already served. The Spanish press were talking earlier in the week about a quarter on quarter contraction of 0.3% in Q2, and it is hard to see any acceleration of the economy in Q3. Pedro Solbes, when questioned explicitly by Punto Radio on the possibility that whole year growth for 2008 could turn negative replied diplomatically "It's not my feeling at the moment", which means basically that it might well turn out to be the case.
If this expectation if fulfilled then Paribas may have to revise their latest forecast slightly (see above link) since - in what is really an excellent general analysis - they pencil-in the recession to start in Q3 2008 and then move on to anticipate a contraction in the Spanish economy of 0.75% in 2009 (although as they freely admit all the risks here are skewed to the downside). My own personal call at this point is that the recession may well have started in Q2 (we will soon know) and that the contraction in whole year 2009 will be over 1 percentage point. Further than that I am not willing to go at this stage, since it all depends, and in particular it depends on whether or not we get a nasty "event" or series of events which send the economy hurtling out of the "hard landing" bracket and into the "melt down" one. It is because I strongly believe we be should doing everything we possibly can to avoid that eventuality that (and not continue to languish under our blankets with a heavy dose of the Artemio Cruz syndrome) that I am writing this post now.
Before continuing, however, I should point out that even the Paribas idea of negative growth in 2009 is still very nonconsensual, despite the widespread pessimism which currently surrounds the Spanish economy. The consensus economic survey for June gives a median 2009 growth forecast of 1.5%. The lowest forecast in the survey is 0.4% but most are grouped in the range 1.0-1.8%. Maybe the consensus will catch up with the curve in due course.
So what would be my justification for making such an apparently gloomy forecast? Well as I argue in my Artemio Cruz piece, and as Paribas re-iterate in their study, this is no ordinary crisis. It take place against a background of a severe credit crunch which affects the entire financial sector, in a country with an enormous external deficit (CA deficit over 1o% of GDP and rising) and which has a strong external energy dependence, and at a time when energy prices have been rising sharply. All of this is bound to put a very strong pressure on internal consumer demand, and as a knock-on impact on investment spending. At the same time slowing growth globally, and in the EU and eurozone economies in particular, makes for a very difficult external environment where increasing exports (even assuming Spanish export prices were currently competitive, which they aren't) becomes difficult, if not well nigh impossible.
Serious Structural Distortions
So let's take a quick look at some of the underlying structural issues. In the first place both Spanish households and corporates are extremely highly leveraged at this point in term in terms of their outstanding debt obligations. The levels of debt to GDP are really extraordinarily high when compared with their other eurozone peers. Click to enlarge:
So how did Spain get into this rather precarious situation? Well I don't think you need to look too far. As can be seen in the chart below, Spain effectively had negative interest rates throughout the entire period between the start of 2002 and the autumn of 2006. That this state of affairs was produced in the very earliest years of the history of the eurozone was indeed, in my opinion, truly unfortunate, since it meant that inflation expectations had not had time to be "steered down" by a central bank track record, and a very widespread reaction to what were generally perceived to be derisory interest rates for savers was to withdraw money from longer term deposit accounts and to place it in what was considered to be the safest of safe inflation hedges: property. Thus began what may well turn out to have been one of the most serious property bubbles in recent history.
Click to enlarge:
The situation was also doubly unfortunate, since the ECB along with other central banks had lowered interest rates in an attempt to support economic weakness produced by a drop in stock market values produced by the collapse of the internet boom. In Spain's case however, the excesses caused by the internet boom never really had the opportunity to unwind, since as one boom ended, another one simply got going in its place. This effect can be clearly seen in the chart for long term quarterly growth produced below, where we can see that following the 1992/93 recession (and up to Q2 2008) Spain simply hasn't had one single quarter of negative growth - that is during 15 years. Hence the legend of the Spanish economic miracle was born. But as with all legends, we should also really be asking ourselves what the reality was which lay behind, since as we can now see, the absence of recession - and in particular the absence of recession in 2002/03 - simply means that we now have a lot of extra "distortion" lying out there and waiting to be "corrected"(the waste-pipes were effectively never flushed, which is why we are now faced with such a peculiar smell emmanating from the sewage system). This would be the main reason why I would argue that what we cannot now expect is a relatively smooth "return to trend" in 12 to 18 months time, since Spain has effectively been "off trend" for some six or seven years now, and the magnitude of the excesses (10%+ CA deficit, 5 million immigrants in eight years, corporate indebtedness pushing the 120% of GDP mark etc etc) is prima facie evidence for this. So even in the best of cases we are almost certainly now facing a significant period of negative and then very low headline GDP growth. But we may not be lucky enough to get away from all this with a simple best case scenario.
Click to enlarge:
The last piece of structural evidence I would offer in this post refers to the CA deficit situation. Since I deal with that reasonably exhaustively in the Artemio Cruz piece, I will only refer to one item here: the deteriorating balance on the income account. Click to enlarge:
Now this is important in my opinion. It is important since obviously any of the remedies to Spain's short term financing problems imply borrowing money (in some way, shape or form via the support which is offered by belonging to the eurosystem). Spain needs one of those proverbial "bailouts", even though since Spain does not have its own idependent currency this position is somewhat masked by the fact that everything is denominated in euros. But debts incur interest, and the more you borrow, the more you effectively have to pay, not only in capital, but also in interest. And if Spain country risk rises sharply in any way - as some analysts are suggesting it may have to - well then what is already a serious problem is only going to become a more serious one.
So where are the risks? Well I think it is no simple accident that Martinsa-Fadesa has been the first major developer to go, since a very large part of their portfolio is composed of land. According to press reports Martinsa Fadesa had land totalling 28.67 million square metres, 41 percent of which is outside Spain (and 50% of which is not "zoned", that is it is without the necessary premission to build). They also have a stock of more than 173,000 newly-built and unsold properties (again by no means all of these are in Spain). Now land is going to be a very important component in this whole "correction", since a lot of land (as we can see) has been accumulated with the objective of construction, and much of this land may now become virtually worthless. And land prices are already falling faster than house prices. Data from the Ministry of Housing shows that land for building fell to 251 Euros/m2 in March, a 7.7% drop when compared with March 2007. Land prices had fallen for 3 consecutive months by March with the average cost of land in Spain now being back somewhere around where it was at the end of 2004.
So I would say this is one of the first issues the Spanish government needs to tackle, and quite urgently. Frankly I can see little alternative to having the government intervene and take this land off the books of those who are holding it - not at market prices, they can handle some sort of "haircut" - but I don't think the government should be sitting idly back and watch one developer after another simply fold, since the end result of this is that the problem then moves over into an already overstretched banking sector.
Which brings me to my exhibit one: Japan land prices.
Click to enlarge:
One of the key features in Japan's ongoing battle against deflation has been the way in which land prices were simply allowed to fall after the property bubble burst in 1991. The above chart shows the sharp rise in Japanese land prices from 1986 to 1990 - a period during which they more than doubled - and how they subsequently fell - albeit more gradually — by roughly two thirds from 1990-91 to 2005. Although urban land prices have turned up slightly in 2006, land prices still continue to fall elsewhere, and of course we still haven't seen how the latest construction "bust" in Japan is going to leave things. Unsurprisingly, residential construction has remained virtually dormant in Japan over this entire deflationary period.So the question is, what is to stop this happening in Spain. I would be grateful to anyone who can present me with a reasoned argument as to why it won't happen. Meanwhile the risk is evidently there.
The Builders In General
Obviously even if the most immediate and pressing problem is arising in Spain in the area of land prices, the rest of the houseing related construction sector will not be far behind. This is a problem that is simply waiting to happen. According to the latest data from the Spanish housing ministry, average Spanish property prices fell by 0.3% in the 3 months to the end of June, but they were still 2% up on prices in Q2 2007, a factor which is leading many to question the reliability of the Spanish data (one more time Artemio Cruz strikes, since Spanish institutions are far from swift in responding to problems, and prefer to deny that they exist). One explanation for the present situation may be that prices are being measured in terms of the initial asking price and not the final selling price. Whatever the explanation prices are certainly set to tumble, and even the the G-14 developers’ association, traditionally a staunch of the immobility of property values, has had to admit that new build prices have fallen by 15% in the last 6 months alone, while Cue Mariano Miguel, ex-president and present board member of the much troubled developer Colonial is already predicting a fall in the region of 25% to 30% over the next two years. And new property in Barcelona (which is where I live) is now taking ten times longer to sell than it was only a year ago, according to real estate consultancy Aguirre Newman.
Meanwhile we learn from Jose Luis Malo de Molina, director general at the Bank of Spain (and speaking at a recent conference in Valencia) that the number of new homes which will be completed in Spain in 2008 will beat all previous records (I said this was a system which was slow to react), simply piling one more house after another adding on the glut of newly built homes that are already idly languishing on the Spanish market. Muñoz's explanation for this phenomenon is simply that “the real estate sector can’t turn around quickly, it works in the medium and long term, so this year the properties started at the end of 2005 and beginning of 2006 will be completed, which means the number of new properties on the market will hit an all time high.” As I say, "just in time" may be an idea that has entered the heads of the more agile companies like the textile consortium Inditex, but most of Spain is a very, very long way from being able to offer an agile response. On the anecdotal front, a friend of mine recently went to visit family homes in the North West of Spain. In Vigo he spoke to the owner of a brick factory, and in Leon someone who had a quarry. In both cases production was continuing (there is simply no on/off switch here) but the inventory already had piled up to be sufficient to satisfy normal requirements for 2009, and of course, in 2009 requirements will not be normal, since housing starts in 2008 have collapsed to a forecast of below 200,000 (down from 600,000 plus in 2007).
At this point estimating the volume of unsold housing in Spain is really a question of "its anyone's guess" rather than a matter of scientific fact. The number 1 million is popular, but I suspect this is more a question of giving an easily managed fact than one which is measuring the houses one by one. The same applies to the estimates for the level of likely fall in property values. All we can safely say at this point, I think, is that the number in both cases is large.
The big question for our current concerns is who is exposed to the risk on all this, and the answer to that question is a lot simpler: Spain's already cash-strapped banking system.
One common estimate of the exposure of the banks to the builders would be somthing of the order of 300 billion euros - this is the opinion of Spanish analyst Inigo Vega at Iberian Securities (and it is one I more or less share). So we could say we have something in the region of 20% to 25% (or more) of Spanish annual GDP in play here.
Bank Exposure Through Mortgage Backed Securites
To this second order exposure of the banking system to the construction sector alone (and remember, through the negative impact of all this on the real economy, we never should lose sight here of non-construction corporates, remember all that indebtedness in the first chart) we need to add the exposure of the banks to the cedulas hipotecarias, which alone run to something in the 250 to 300 billion euro range (to which we need to add, of course, other classes of more conventional mortgage backed-securities ), then we are talking about an injection into something of over half of one years GDP in Spain.
According to Celine Choulet of PNB Paribas the proportion of mortgage-backed securities in the broader sense of the term (ie including cedulas and MBS) represent around 37% of outstanding mortgage loans in Spain. She also estimates that asset backed securities held by non-residents constitute may amount to as much as 81% of the total securities issued.
Outstanding home loans (for purchases and refis) represent a substantial percentage of the Spanish banking institutions’ balance sheets (21.5% of total assets and 35.6% of total loans to the non-financial private sector in the second quarter of 2007). In the second quarter of 2007, outstanding home loans amounted to 589 billion euros, 56.4% of which were distributed by cajas (29.8% of their assets), 37.2% by commercial banks (15.4%) and 6.4% by mutual institutions (30.9%).
If we add together home loans and the financing of real estate sector (construction and property services), the overall exposure of Spanish credit institutions has increased significantly over the last decade (37% of assets in the second quarter of 2007, 61.5% of total loans to the non-financial private sector). Exposure of Spanish banks to the real estate sector exceeds, both in level and in growth rate, that of US, Japanese and British banks. In total, in the second quarter of 2007, cajas (49.7% of assets, 70.5% of loans) and mutual institutions (46% and 56.3% respectively) were almost twice as exposed as commercial banks (28% and 55.2% respectively).
According to Choulet - and just to take one example - in 2006 total new funding to the Spanish mortgage market reached 201.3 billion euros, of which 88.3 billion took the form of covered bonds (representing 43.9% of the total of mortgage securities market) and 113 billion was in mortgage-backed securities (56.1%).
And remember the cedulas all need to be "rolled over" in the next few years (with a big chunk coming up between now and 2012. And the problem starts this autumn. According to an article in the Spanish daily El Pais at the end of June the Spanish banking sector needs to raise 62 billion Euros before the end of this year just to rollover what they have coming up on the immediate horizon.
What does all this add up to? Well, to do some simple rule of thumb arithmetic, just to soak up the builders debts and handle the cedulas mess, we are talking of quantities in the region of 500 to 600 billion euros, or more than half of one years Spanish GDP. Of course, not every builder is going to go bust, and not every cedula cannot be refinanced, but the weight of all this on the Spanish banking system is going to be enormous. Banco Popular is the most visible sign of the pressure, and their shares have already dropped by 44% this year, and by 7.9% on Tuesday alone (they were the listed bank which was most exposed to Mrtinsa Fadesa).
So it is either inject a lot of money now - more than Spain itslelf can afford alone - or have several percentage points of GDP contraction over several years and very large price deflation - ie a rather big slump - in my very humble opinion. And it is just at this point that we hit a major structural, and hitherto I think, unforeseen problem in the eurosystem (although Marty Feldstein was scratching around in the right area from the start). Since if there is to be a massive cash injection into Spain's economy, who is going to make it? Spain alone will simply crumble under the weight, and it is evident that the problem has arisen not as the result of bad decisions by the Spanish government, but as a result of institutional policies in Brussels and monetary policy at the ECB. And yet, the Commission and the ECB are not the United States Treasury and the Federal Reserve, no amount of talk about European countries being similar to Florida and Nebraska is going to get us out of this one: it is going to be step up to the plate and put your money where your mouth is time soon enough. Yet, cor blimey, we are still busying ourselves arguing about the small print on the Lisbon Treaty.
Demographics and Construction
The third major area of risk I would like to highlight today relates to the problem of demographics and construction. PNB Paribas (see initial link) put it like this:
"With United Nations population projections pointing to growth of only 300k per year on a ‘high-population’ variant for 2010-15, housing starts could fall considerably further. Hence the risks to our central forecast of 30% off housing investment by end-2009 are to the downside. The correction could be more rapid than expected. If not, it is likely to persist into 2010. Our forecast has housing investment converging to levels consistent with relatively strong population growth. A weaker population assumption or some undershoot of the ‘equilibrium’ level would lead to a worse outcome."
Basically, I think the big topic here is the coming rate of new household formation. And here it is worth remembering that while the countries most affected by the property-driven credit crunch in the EU would appear at this point to be Spain, Ireland and the UK, the UK is rather different from the other two, since while usebuilding grew by 187% in Spain between 1996 and 2006 (and by 177% in Ireland), the equivalent increase in the UK was just 12%. Planning restrictions in Britain meant fewer homes were built which may also explain why house prices in the country have almost doubled, in real terms, since 1999 despite the comparatively low percentage of new builds (this would bring us back to the huge zoned and non zoned lang overhang in Spain, and what the dynamics are that produced it). That is, while the UK can to some extent offset the impact of the crisis by increasing homebuilding at some point (to house, for example, all the extra people from Poland and other parts of Eastern Europe), in Spain and Ireland the problem is going to be very different, since they have to sharply reduce housebuilding.
So what are the sources of new household formation in Spain? Well basically they are threefold: natural population development, migration, and second homeowners from the north of Europe. Now if we start with the question of natural population evolution in Spain, it is clear that ex-migration the Spanish population is virtually stationary at this point - with an average annual increase of a mere 30,000. But what matters in housing terms is not so much the size of the population as its age structure, and here we don't need to go to the refinement of looking at longer term UN projections (high, median or otherwise) because in terms of Spanish property from now to 2020 (at least in terms of natural population drift) the deal is now done (or rather the goose is now cooked). A quick glance at the US Census Bureau IDB population pyramid for 2000 should make this abundantly clear.
Click to enlarge:
What we can effectively see is that in 2000 (and please click on the image if you want a better look) Spains three largest historic cohorts were in the 25 to 40 age group. But if we mentally fast forward as far as 2015 we will see that the aggregate size of the cohorts in this age range is very significantly smaller, and if we fast forward again to 2020, we will see that what we have are the three smallest cohorts in the last forty years. And from here on in we only go down and down - talk about absence of sustainability!
So we are left with North Europeans is search of second homes and migrants to offer some support to Spain's rapidly crumbling housing sector in the coming years. Well on the North Europeans front the picture doesn't look exactly promising, since the bulk of the buyers in recent years have been British (Britons own an estimated 500,000 to 700,000 properties in Spain), and they are already having their own problems, plus the fact that changes in the value of the GBP and interest rates mean that affordability is becoming an issue, an issue to which you have to add the attraction of properties whose prices may now be set to seriously deflate, and over a significant number of years.
Indeed, according to Manuel Gandarias, president of the ‘Live in Spain’ holiday-home developers’ association, sales of holiday homes in Spain are now down by 50% from the peak “In recent years between 120,000 and 125,000 holiday homes were sold each year, this year it will be half that,” he is quoted as saying. And of course it isn't only the cost of the home that has been going up, it is the cost of servicing the debt the home brings with it. Josep Suárez, director at Solbank in London estimates that the combined impact of rising Euribor rates and the appreciation of the Euro against the pound (15% in the last 9 months) means that mortgage payments for Britons with mortgages in Spain are now 25% higher than they were a year ago.
So the outlook on the North European second home market doesn't exactly look bright either, which leaves us with the migrants. As is now generally well known, Spain's population has increased dramatically in recent years - from around 40 million in 2000 to around 45 million in 2008 - and this increase has been almost exclusively (natural increase is no more than a quarter of a million) the result of huge inward migration. Click to enlarge:
Basically the future of all these migrants is now deeply uncertain. I would even say that losing the migrants constitutes the most important of all the downside risks to the Spanish economic crisis for the impact it will have on urban rents and mortgage delinquency in the short term (since many of the migrants have bought flats), and for the cosnequences for Spain's housing market and pensions system in the mid term.
Is There A Rescue Plan Available?
Basically, and on the basis of all the above, I would like to put forward a five point "rescue" plan for the Spanish economy. This would look something like the following:
1/ Set up a national land agency, to buy up land and to convert to other uses (agriculture wouldn't be a bad bet where possible given present food prices). This to include the proviso that such land could never again be zoned.
2/ Buy out and close down builders the bankrupt builders as part of a general restructuring programme such as the one which was developed for the shipyards and the mines.
3/ Buy up and burn ALL outstanding cedulas hipotecarias
4/ Establish a programme to help immigrants in difficult circumstances, and offer training etc to prepare for the future
5/ Restructure all existing mortgage contracts - which will involve every one paying more - in order to put mortgage financing in Spain back on a sound footing.
OK, I warned you. None of this was going to be popular. And non of these propsals should be consider as carved in stone. Better ones could well be put forward, but in the absence of anything credible in the way of alternatives I am putting them forward. As I said at the start, there is no point in agreeing to have your own throat slit just to see people you don't like have theirs slit first.
It is very important that some form of "corta fuegos" (fire break) is put in place, and put in place now, otherwise the whole of Spain could very easily burn down the same way the Liceu opera house did here in Barcelona, just because some chump decided to do on-stage soldering with the safety curtain up! Risk sir, there's no risk here. It's all as safe as houses.