Friday, July 13, 2007

Latvian housing bubble, part 4

Latvia may be having a housing bubble and the prices have started dropping slightly (4% in the last two months). How far could they drop?

First, Latvia has a lot of people who would want a bigger apartment. According to this study by the Bank of Latvia, the levels of apartment space per person in Riga are lower than in both Tallinn and Vilnius. And even if the present construction boom continues at the current rate, it will take 8.6 years to reach the current level of Tallinn.

The problem is the prices that people are willing to pay. According to the same study, the price of 1m2 in Riga is 3.5 times the average monthly salary. That is more than in any other European country for which they have data. (The next most expensive country is Spain, with 1m2=3 times the monthly salary.) They also counted the number of people who need a bigger place and are able to qualify the loan at the present prices. The conclusion? If not for people who are buying property for investment or to rent it out, Latvia would have run out of mortgage applicants in 6 months after the study. (Maybe slightly more, if one includes people with "unofficial" incomes which are not declared to tax service, but not much more.)

The other key variable is the credit availability. As I wrote, the home loans are mostly funded by money borrowed from abroad and Latvian banks have been borrowing abroad a lot. The continued willingness of foreign banks to lend ever-larger amounts is questionable. I suspect this is, actually, the most important constraint.

About 50% of loans are funded by local savings. The remaining 50% are borrowed from abroad. If the inflows of credits from abroad decrease by 25-50%, that means that the amount of money available for credits decreases by 12.5-25% of what it was at the peak of the lending boom.

In the first approximation, I would guess 12.5-25% decrease in prices based on that. It may be less because some people would withdraw their apartment from sale, instead of selling them for less. On the other hand, the new housing projects are being finished now and when they are ready, the number of apartments will increase and this will press the prices of existing apartments down. (This is already happening in Estonia.)

This is all my guesswork. (Do not regard this as an investment advice!) I'd love to hear what other people think.

4 comments:

Edward Hugh said...

Hi LA,

Great work. Just keep doing this, at the end of the day you could have quite a work of art. A complete record, for posterity.

Now.

"the levels of apartment space per person in Riga are lower than in both Tallinn and Vilnius."

This seems to be a typical situation, a supply side constraint which produces a spike in prices.

I would say it is important here to differentiate between those countries which have fueled the boom with migrants - Spain, the UK, and Ireland - since these countries can use government finance to boost supply artificially (given that affordability in the open market is now a general issue in these countries with the rise in interest rates).

This unfortunately can't happen in Latvia since, as you will have noted with Ammunia's latest statement backing the original IMF proposal, with a 4% budget surplus being the only real policy instrument they have right now, they just can't get into subsidising public housing.

So construction could well power to a halt, and then the underlying problem simply remains, unless of course the demand side corrects through the tragedy of outward migration.

"The problem is the prices that people are willing to pay."

More than what they are willing to pay, it may be what they have been able to pay, given the availability of cheap credit, an availability which may be about to dry up (and we may even see some effect from the US sub prime bust here). Risk appetite among lenders WILL be down.

"The remaining 50% are borrowed from abroad."

Well, there are also remittances to think about. My feeling is that these are a significant part of the story. The Polish national bank just suggested that Poles abroad had sent home around a billion euros in the first quarter 2007 alone. All this money has to go somewhere, and my guess is that a lot of it goes into housing. First generation migrants are normally more savers than consumers, and housing is, after all, generally seen as a form of saving, especially in times of inflation.

"This is all my guesswork. (Do not regard this as an investment advice!) I'd love to hear what other people think."

Well this is all very hard to say at this point. I'm afraid I am not optimistic. On top of the credit tightening you need to think about feedback mechanisms in domestic consumption. These are of two kinds:

the wealth effect
the decline in construction itself

Now funnily enough the second may not be so important since you are so short of labour that people may simply be able to move sideways.

And then on the wealth side, remittances will keep coming, and may even accelerate if even more people leave in search of work. So this can underpin prices. The big question is how many of them will ever actually come back?

Recent surveys among Poles in the UK are not very encouraging on that count.

This is all sufficiently complex that it becomes very hard to actually call. All I can suggest is that we watch and wait. But remember, with the peg, and the very high euro, the competitiveness issue is going to be a very important one.

And if they ditch the peg? Well look out is all I can say.

Latvian abroad said...

Thanks for the comment! It's insightful, as always.

On remittances, the amount that Latvians sent home in 2006 was approx.
270 mln lats/380 mln euro/2% GDP.
In per capita terms, that might be even more than in Poland (whose population is 17 X Latvia).
It's order of magnitude less than the credit inflows, though. The increase in home loans over last 12 months was:
1.88 bln lats/2.7 bln euro/14% GDP.
So, remittances are certainly a contributing factor but not as big one as cheap credit. And they won't be enough to compensate for the disappearance of cheap credit (partially, maybe, but not fully).

Edward Hugh said...

"On remittances, the amount that Latvians sent home in 2006 was approx. 270 mln lats/380 mln euro/2% GDP."

Interesting.

"So, remittances are certainly a contributing factor but not as big one as cheap credit. And they won't be enough to compensate for the disappearance of cheap credit (partially, maybe, but not fully)."

No, this is clear, but we still have to see how much it will actually be possible to squeeze cheap credit. Those with an interest in promoting construction activity will be looking to find more sources. Maybe they can, maybe they can't. The sub-prime bust in the US you would have thought would be making people nervous about this kind of thing.

Otoh there are other tendencies, and in particular the rise of funding from Japan. I don't know if people in Latvia will start to leverage this possibility. I do notice though how people here in Spain are starting to mention yen denominated mortgages as a way round the obstacle being offered to Spanish construction interests by ECB monetary policy right now.

And I can't help also noticing how Yen denominated mortgages started to take off in the UK back in 2005, just when Mervyn King tried to squeeze the housing situation there. I have been most struck by the resilience in UK housing despite the rates rises, and I am looking for scapegoats. Yen mortgages and new Polish residents undoubtedly form *part* of the picture.

"And they won't be enough to compensate for the disappearance of cheap credit"

No, but if we look at the orders of magnitude, they can offset (or should I say frustrate) the government surplus issue being proposed by the IMF (and now Almunia). It all depends on how this flow of remittances evolves.

So while normally you would expect an economy with such extreme imbalances as the Latvian one currently has to fold like a pack of cards once someone squeezes enough, there are grounds for thinking that this situation in ever so many ways is far from normal. So we need to keep an open mind and see what we can learn.

Interesting point you made on DM about migrants (btw the graph link doesn't work), this suggests that some labour market easing can come from this quarter. I certainly hope so. Obviously there is going to be a slowdown, the question is how big it will be.

But then of course we are left with the external imbalances. If the present levels of costs and prices hold there is a big competitiveness problem. In the longer run this can - as in Germany - be resolved by internal, structural, productivity adjustment, but this needs time to work (years) and Latvia has to get from here to there. So in the short run there is bound to be pressure on the Lat peg at some point.

Going back to the CIS migration thing, it is clear that this can help, but as I am trying to argue with the Economist, it can only help if it is seen as a way to buy time to do other things. I still think that Poland puts a sort of outer limit on the situation here, and that when Poland reaches unemployment in the Latvia/Lithuania region one game will be over and a new one will have to begin. Doubly so, since you have to think about Western Europe's needs here. Germany is going to need labour, now where is it going to come from?

ste tomkinson said...

I have just been looking at property in sharm el shake, This was helpful thank you