Falling Swedish Housing Prices Again

I think it's about time I updated the English version of my Swedish housing bubble diagram and analysis from September 2007.

The following is based on articles from my Swedish language blog from the past year or so.

At the end of 2010 real house prices in Sweden fell quarter over quarter for the first time since the first quarter of 2009. According to Statistics Sweden's "Fastighetsprisindex småhus" and CPI, inflation adjusted house prices fell from the third to the fourth quarter of 2010. The falling prices have continued during the first quarter of 2011. See diagram below for prices in the whole country and in the Stockholm area.

For comparison I've also included real wages. The Swedish housing bubble is credit driven, as you can see if you compare housing prices, wages and household credit. In 2007 I thought we had seen the top, but during the financial crisis of 2008 the Swedish central bank, Riksbanken, took interest rates down sharply, and since Sweden was relatively mildly hit by the crisis the Swedes started bidding up housing prices even more using borrowed money at low interest rates.

The question now is - have we seen the real top in the Swedish housing market this time? I find it hard to see any factors that could inflate the Swedish housing bubble further, so this time I definitely call it a top.

There are always "fundamentalists" who argue that there is no housing bubble in Sweden, citing for instance supply and demand showing that there is not enough housing in for example Stockholm. However, housing price increases in Sweden are not limited to the large cities. Percentage-wise prices have even increased more in some smaller towns. Sweden also has among Europe's largest dwelling area per capita, even compared to neighbours such as Finland.

What the "fundamentalists" also miss is that this is really a household credit bubble, which happens to manifest itself in housing prices. It also manifests itself in household consumption, which continues unabated in Sweden, though there are recent signs of a softening. This of course drives the strong Swedish GDP increase, which however is largely driven by credit-fuelled household consumption. If household credit expansion should stand still, Swedish GDP will no longer increase.

Now it's not just Sweden that still has an unpopped housing bubble. Our neighbours Norway and Finland also do. Denmark's bubble popped in 2007, however, which has caused the demise of nine smaller Danish banks, but has so far left the large banks relatively unscathed.

Norway's bubble is bigger than Sweden's, and Finland's is smaller, using 1993 as the base year. All the Nordic bubbles are larger percentage-wise than the American one. One cannot help but wonder what will happen to the Norwegian bubble, considering that Norwegian oil production has been falling for about ten years now.

Also read the Australian blog Macrobusiness' recent analysis of the Swedish housing bubble.

Western European Banks' Exposure to Eastern Europe

This article is a translation of an article on my Swedish language blog.

Last week I wrote about Eastern Europe's loans from Western European Banks (in Swedish). Now I've taken the latest figures from BIS and made graphs to illustrate how the situation is. Please note that the figures are only for claims from banks. There may be other lenders too, but it's not as easy to find statistics for them.

First I'll present an illustration of which Eastern European countries have borrowed most. The colours indicate which in Western European countries the they have borrowed from belong. Click the graph for a sharper version.This graph shows clearly that we don't need to worry too much about Western Europe's banks if Macedonia, Moldova, Montenegro, Belarus, Albania, Bosnia or Serbia should be hit by a collapse of the economy and currency, since their loan volumes are relatively small. The other countries, however, are potential threats to Western Europe's banks. This becomes especially clear when one Western European country has a large proportion of the lending to a certain Eastern European country, e.g. Sweden to Estonia, Latvia and Lithuania.

Huge figures have been floating around when it comes to the exposure of Western European banks to Eastern Europe, but the total sum of bank lending is "only" slightly more than 1.4 trillion dollars, which should be quite close to the actual figure. This is still, however, a very large figure.

Now I'm going to look at it from "the other side", i.e. divided per lender country instead.Here you can see that Austria is at the top of the risk list for lending to Eastern Europe. As for Sweden we can note that the major part of the lending is to the three Baltic states. Since these three economies risk following each other, Sweden is much more exposed to a single risk factor.

However, absolute numbers don't often say that much. If Italy lends a bit over 200 billion dollars that's a larger share of the country's GDP than if Germany lends the same amount. So below I compare the banks' lending to Eastern Europe to the GDP of the countries (taken from the 2008 GDP estimates from the CIA Factbook).Then you suddenly see what a huge risk Austria is taking! Their banks have lent roughly 64% of the country's GDP just to Eastern Europe! Even a mild crisis in Eastern Europe would thus probably suffice to topple Austria's banks - we don't even need a serious crisis. The biggest Austrian claims are on Czechia, Romania and Hungary. At least Romania and Hungary are already on the problem radar...

By Austrian measures the other Western European countries have taken limited risks towards Eastern Europe, even if they definitely aren't negligible. Belgium's banks have lent about 25% of Belgium's GDP to Eastern Europe, and Swedish banks about 20% of our GDP, most of it to Estonia, Latvia and Lithuania.

Let us also take a look at which Eastern European countries have borrowed the most in relation to their GDP, and thus have taken a large risk. This is according to my usual identification mechanism for potential problems in this crisis. Large loans = large problem risk.Here you can see that Croatia and Estonia are in a risk class of their own. Their claims from foreign banks are more than 120% of their GDP. But Latvia, Hungary, Czechia, Slovakia, Lithuania and Bulgaria are also risk factors that cannot be ignored (the nominal sum of their borrowing according to the first graph in this article also plays a part here). Add to this the fact that GDP for many Eastern European countries is now falling rapidly, and we will quickly get even bigger loans-to-gdp ratios.

Part of the risk is also the currency risk. A large part of the loans from Western to Eastern Europe are denominated in euros or Swiss francs. This causes payment difficulties when the local currency sinks against the euro and/or the franc. Unfortunately I don't have any figures for how much of the loans are denominated in foreign currencies, but we can take a look at how some currencies have moved against the euro since the end of August last year.Here you can see that Ukraine's hryvnia and Poland's zloty have fared worst. We have all heard about Ukraine's problems, but there has not been much mention of Poland during this crisis. Hungary, Romania, Russia and Czechia have also been hit by falling currencies, though not as badly.

To conclude, we can see that the banks of several Western European countries have taken large risks towards the now faltering economies of Eastern Europe. There is an immediate risk that for example a currncy and/or economic collapse in one of the risk countries I've presented above could topple e.g. Austria's banks. This in turn would probably give serious repercussions for the whole euro area (Austria is part of it). As I see it it is only a matter of time before something drastic happens. Exactly what will happen is hard to predict, but Sweden is definitely also in the risk zone.

In this context it is interesting to note that the bank crisis of 1931 had its origins in Austria, where their then largest bank Creditanstalt went bankrupt in May of that year. Will we get a replay?

P.S. I recommend the following two articles by Ambrose Evans-Pritchard at the Telegraph for those who want to read more about the risks from Eastern Europe:
Failure to save East Europe will lead to worldwide meltdown
Eastern European currencies crumble as fears of debt crisis grow

Swedish and Danish Kronas and Interest Rates

This is a partial translation of my original article in Swedish.

Sweden and Denmark have gone different ways when it comes to currencies. Both countries have kronas, as you probably know, and have not joined the euro. But while the Danish krona is pegged to the euro through the European Exchange Rate Mechanism ERM the Swedish krona has no peg. This can be seen from how the Swedish krona, being a small currency, fluctuates much more than the euro against the dollar.As you can see from the graph above one paid about 9.30-9.50 Swedish kronor for a euro until the beginning of September 2008. When the crisis then gained momentum the krona fell against the euro so that one at times had to pay more than 11 kronor for a euro (today we're at 10.76 kronor per euro). This means that the krona lost about 13% against the euro, just because we have a small currency, which is not attractive to investors in a crisis. Theoretically a weak currency should be good for our export industry (Sweden runs a current account surplus) but in a global crisis where world trade is generally diminishing one can question how much this is worth.

The Danish krona however has been stable around 7.45 DKK/EUR all the time - lucky them, one could believe. But the Danes have been hit by something else - look at the diagram below for the rates of the Danish central bank (red), the European Central Bank (blue) and the Swedish central bank (green):Danmarks nationalbank had to raise their rate at the beginning of October to defend the value of the Danish krona against the euro, when ECB and Sweden instead lowered their rates to cushion the effects of the financial crisis. Even though the Danish central bank rate has been lowered in three steps since then, it is still a good deal higher than the ECB rate, which has also been lowered during this period. The higher interest rate climate in Denmark can also be seen when you look at the interbank rates - Danish 3 month CIBOR was at 3.9317% today, while Swedish 3 month STIBOR was at 2.023%. A rather hefty difference, I would say. So this is the price Denmark pays for its stronger currency.

Now voices are being raised in Sweden for us to join the euro. The Swedish small enterprises association's leader Anna-Stina Nordmark Nilsson and professor of economics Harry Flam have written an article each in Sweden's two largest newspapers where they argue for joining the euro. Well, it might not be very good timing to join the euro now, when it seems the euro co-operation itself might break.

Anyway, Sweden cannot join the euro straight away, since a country's currency must be part of the ERM for two years first according to the rules. So the Swedish euro friends will have to wait for the continuing onslaught of the financial crisis before they get what they want.

Swedish Housing Still a Bubble

My view is still that Sweden is in a housing bubble, like most western European countries. Signature "ju" asked me on a forum why I don't compare to incomes, which have risen in Sweden over the last years. So I did some more number crunching, and here's an improved version of my Swedish housing bubble graph, with a green line for inflation adjusted mean income. I've also included housing price data up to Q2 2007 and trend lines (thanks to "Kons" for suggesting trend lines).
(Data source SCB. FPI=Fastighetsprisindex småhus, Mean Income=Sammanräknad förvärvsinkomst. 1975=100 for housing prices, 1999=100 for incomes)
As you can clearly see, housing prices in Sweden have risen much faster than the average income (I only have income data for 1991-2005), so while inflation-adjusted incomes have indeed risen, we are still in a housing bubble. How far will housing prices fall?

• A fall down to the trend line is about 30%.
• A fall down to the average for 1975-2003 is about 45%.
• A fall down to the lows of the early 1990's is about 55%.

This is the whole of Sweden, and obviously the bubble is worse in "hot" housing spots. So I've made a corresponding graph for housing prices in the Stockholm area (thanks to "Kons" for suggesting this).
As you can see the graph is slightly more "bubblish" than for the whole of Sweden.

• A fall to the trend line is about 30%
• A fall to the average for 1975-2003 is about 55%
• A fall to the lows of the early 1990's is about 65%!!

Many people will of course say "it can never happen, there will always be a strong demand for housing in our capital city", but I boldly predict today that inflation-adjusted housing prices in Stockholm will probably on average go down by 65% over the next few years. Be warned, and don't forget where you heard it first. And I might even be optimistic. Considering the excesses on the upside, we could very well have similar excesses to the downside when the bubble finally bursts. And if you don't believe it could happen, don't forget that history has proved that all bubbles eventually burst, even though few believe it is possible when on the upward slope of the bubble.

Now housing bubbles burst slowly, so the process will take a few years to play out. I expect "experts" to call a bottom in housing many times during the coming years, but prices will still keep on falling.

Now what effects will such a large fall in housing prices have on the economy? Large effects, of course. Many people will be stuck with loans way bigger than the value of their home. But maybe we should ask the question the other way around - what economic events could trigger such a large fall in housing prices? Read what I have previously written on this blog about potential dangers in the world economy, and you will get some hints about what might cause such a powerful recession.

Speaking of housing bubbles, the top economic advisor of the Spanish prime minister came up with some unbelievable quotes last week:

A residential real estate slump in Spain, where prices have almost tripled since 1997, is "unthinkable," the top economic adviser of Prime Minister Jose Luis Rodriguez Zapatero said. [...]
"To talk about severe adjustments or a meltdown in prices is ridiculous," Taguas said in response to reports pointing to an end of the Spanish real estate boom. "That sort of crisis is unthinkable." [...]
The Spanish banking system is also solid enough to withstand rising financing costs triggered by the fallout from the surge in defaults in the U.S. subprime mortgage market. A run on mortgage- lenders such as Newcastle, U.K.-based Northern Rock Plc or funding difficulties like those at Countrywide Financial Corp. in the U.S. are "unthinkable" in Spain, Taguas said.

Remember those quotes and remind Mr. Taguas about them in a few years' time!

And Europe?

It seems that U.S. financial institutions have managed to export some of their worst tranches of subprime loans to Europe. This can be seen from the size of the interventions by the central banks. On Thursday last week the European Central Bank had to inject €95 billion in liquidity into the markets (about $130 billion), whereas the U.S. FED only had to inject $24 billion. We still don't know the whole map of who sits on what bad credit, but it will be revealed eventually.

It is now obvious that the subprime mortgage mess is not "contained" in any way, and is definitely not contained within the country of origin, the U.S.A.

There has been a lot of news lately about U.S. housing and mortgages, but I expect all this to turn up in Europe too. Housing prices have been rising like a bubble here too, and many people are up to their ears in mortgages, and have no margins in their economy. Inspired by Robert Shiller's graph of inflation-adjusted U.S. home prices, I compiled one for home prices in Sweden. Unfortunately I could not easily find data before 1975, but here it is:
I've taken the FPI (Fastighetsprisindex småhus) divided by KPI (Consumer Price Index) and multiplied it by 1000 just to get some nice round numbers (figures from SCB). You can see that the prices have gone up and down in waves more or less like the U.S. prices from 1975 to 2000. Then, just like in the U.S. since 2000 they have risen above any previous tops, though the bubble is not as big as in the U.S. The situation is similar in most other western European countries - worse in some (notably UK and Spain), maybe better in others.

Now that lending standards have suddenly tightened up, I expect housing prices to fall in Europe too, and soon we will have our own version of the subprime mess. All major Swedish banks have said last week that their exposure to U.S. subprime mortgages is minimal, which I believe. What they do not mention is their exposure to the Swedish mortage market, which could quickly turn sour.

We definitely haven't seen the last fallout from the global tightening of credit. There are lots of pockets of bad credit all around the globe and in all kinds of markets, just waiting to pop up.