The macroeconomic data coming out of Estonia in recent weeks are truly shocking even in the context of the ten percent annual drop in GDP for 2009 that most observers are now forecasting. Perhaps the most evocative number of all is not the 27% year on year drop in industrial output registered in January, but the announcement this week that Estoniaâ€™s registered unemployment rate rose to a record 7.4 percent during the first week in March, with a total of 47,774 job-seekers registering with the unemployment offices, up 3,019 in a week. Of course, for many outsiders these are not large numbers, but then Estonia is not a large country. Still this was the highest number since the Labor Market Board started disemminating data in 1993 (although not as measured by Eurostat, which uses a different methodology). The level was up from 7.1 percent at the end of February and 6 percent in January, although the important thing is not the volume of unemployment, but the rate of its increase.
At the same time it is estimated that nearly 250,000 Estonians are currently living in homes whose market value is insufficient to cover the outstanding mortgage loans which their owners have taken out, making “exposure risk” a growing problem for the country’s banks. During the boom, house sale transactions were commonly financed with a 90% loan to value (LtV) ratio. This is a very dubious practice at the best of time, but in the face of a sharp fall in both house values and wages it becomes well nigh disastrous.
Once boasting one of Europe’s fastest-growing real estate markets, property prices in Estonia fell by a whopping 23% in 2008 (following an 18% increase in 2007) according to data in the latest edition of the Royal Institution of Chartered Surveyors European Housing Review. The RICS tracked 2008 year-on-year house price inflation in 18 West and East European countries, and found that Estonia’s fall was the most substantial in the entire group.
Take, for example, a 50 sq metre apartment bought in the spring of 2007 for a price of around EEK 1.3 mln. This apartment is currently worth around EEK 790,000, but the outstanding loan balance is of the order of EEK 1.1 mln. Should the once proud owners of that lovely appartment now find themselves among those unfortunate enough to be queueing up outside the offices of the Estonian Labour Board and need to sell it, then even assuming they could find a buyer they would not only lose their home, but they would still end up owing the bank EEK 300,000 under Estonia’s “full recourse” lending laws (which are of course very different from those operating in the United States). With an average net monthly salary in the region of EEK 10,000 this means that the unfortunate ex-property owners would in all probability end up with a debt worth more than two years their total income.
Of course, in this climate buyers are likely to be scarce, and it is more probable that the banks themselves end up with a substantial direct interest in Estonia’s property market. And this would only add to the problem they are already having with overdue loans, which are rising and reached 3.6 percent of total credit in January, according to the most recent data from the central bank which now forecasts bad loans will hit 6 percent before the year is out. Of course, as is by now well know, more than 95 percent of Estonian banking assets are held by Nordic banks, and despite the fact that the banks don’t cease to reassure us that their Baltic operations form a â€œkey partâ€ of their business and that they have a â€œlong-term commitmentâ€ to Estonia, this doesn’t stop them getting downgrades. Swedebank, for example, had its credit rating cut to A1 from Aa3 by Moodyâ€™s Investors Service last month, citing the risk of a â€œsubstantial increase in impairmentsâ€ (read loan defaults and deteriorating asset quality) from the bank’s Baltic operations.
Meantime output and employment simply keep on falling, with Estonia’s industrial production dropping by the most in at least 14 years in January – 26.8 percent year on year, the most since 1995 (following a 22.4 percent slump in December).
Of course, as output drops and people are sent home to remain inactive, the one thing Estonia does have at the moment is a lot of loan offers. Thus the central bank recently announced that they will be able to borrow as much as 10 billion Swedish kroner against Estonian krooni from their Swedish counterpart in an attempt to boost confidence in Estonia’s financial markets. As Riksbank Governor Stefan Ingves said in the statement â€œThe financial systems in Estonia and Sweden are closely linkedâ€. But what Estonia needs is not more loans, and more debt, and people lying around idle, it needs work, and output, and exports to pay off all that debt which has been accumulated. And it is just at this central point that the current solutions are being tested and found wanting.
The Price and Wage Correction Is Too Slow
In order to understand what is wrong with the path on which Estonia has set itself we need to bear fully in mind that the problem is that the country (or its households) have become excessively indebted in relation to the economy’s competitiveness, and the consequent ability to pay. Estonia has a current account deficit, and this does not help things, but Estonia’s problem is not, in the longer run, a simple balance of payments and financial crisis one (against which external loans can of course help), but a problem of competitiveness and the ability to pay off debt.
And even despite the recent sharp fall – almost all of which is produced by a fall in imports and a reduction in living standards – Estonia’s current account deficit was still running at slightly over 9 percent of gross domestic product in 2008 (following the 18.1 percent shortfall achieved in 2007).
Estonian central bank data show an estimated current account defict for last December of 943 million kroons, down from a revised 1.87 billion kroons for November, and from around 3.5 billion kroons in December 2007, but since exports were down 6% year on year in December, it is obvious that the reason for the contraction in the deficit is the 17% drop in imports. Ouch!
Now, as I say, basically the problem here is to restore competitiveness and, although not everyone will be prepared to agree with me, I would argue that the only solution for Estonia is to export its way out of trouble. Given the problems the banking system is having and is about to have, it would be sheer fantasy-land (and very foolish) to imagine we are going to see a return at any point in the forseeable future to consumer credit driven growth (we are talking everywhere about more, not less, regulation), so as Estonians work hard (once they finally get a job again) to pay off their debts and try to save for their increasingly uncertain old age, the only really valid way to try to go for growth is by exporting. Saying that this is not possible, well… this is simply defeatism before you start, and I don’t imagine the Estonian character that way somehow, not after so many years of fighting to gain a hard won independence.
So if you want to export, you have one benchmark to work againt – Germany. And if we look at the chart below, we will see the extent of the competitveness gap which has opened up since 1999. Now Reel Effective Exchange Rates (REERs) are a nice measure of competitiveness, since REERs attempt to assess a country’s price or cost competitiveness relative to its principal competitors in international markets. Since changes in cost and price competitiveness depend not only on exchange rate movements but also on cost and price trends the specific REERs used by Eurostat for its Sustainable Development Indicators have been deflated by nominal unit labour costs (total economy) against a panel of 36 countries (= EU27 + 9 other industrial countries: Australia, Canada, United States, Japan, Norway, New Zealand, Mexico, Switzerland, and Turkey). Double export weights are used to calculate REERs, reflecting not only competition in the home markets of the various competitors, but also competition in export markets elsewhere. A rise in the index means a loss of competitiveness, and as we can see Estonia’s index has risen sharply against Germany’s in recent years.
Well, just in case anyone thinks that the comparison with Germany is not an appropriate one in Estonia’s case, here (see below) is the equivalent chart for Finland, which shows an equally strong loss, and let us remember that the worst year in this sense (2008) is still not included, since Eurostat have not processed the data yet.
And of course, I am only looking at eurozone comparisons here, we won’t enter at this point into the embarassing fact that Sweden and the UK have both devalued sharply in rcent months, as have Eastern EU rivals, Romania, Poland, Hungary and the Czech Republic, as well as non EU rivals like Ukraine and Russia. Really hanging on to the peg blindly in these circumstances is not only foolish, it is ridiculous, and I hardly see how following a ridiculous policy (which for sure won’t work) is going to enhance your credibility, which is what the decision not to devalue was all about in the first place. It won’t even shield the Nordic banks from the slew of incoming defaults.
Now, “plan A” is supposed to involve a very sharp downward adjustment in prices and wages, something of the order of 20% during 2009 and 2010. (Incidentally, talk of a V shaped recovery is misleading here, since the V shaped recovery only comes with a one-off devaluation, say getting the 20% out of the way all at once, and doing it over two years can only bring a U shaped process, as you simply spin the same thing out over two years, think about it, the issue isn’t that hard to see). Anyway, over two years it is, so how are we getting on? Well up to December last year (which is the latest data we have) not very well, since average hourly wages (the key number here) were still up 9.9% in the last quarter of last year, and so this is really another 10% or so to add to the 20% we were just talking about above (based on the 2007 REER). True, hourly wages did peak in Q2 at 78.26 kroon, and were down to 75.58 kroon in Q4 (or by 3.4% in six months), but this was only really taking back some of the excess from H1 2008, and the real hard work is still to come.
But if we move away from wages and take a look at prices, we find the situation is not much better, since while Estoniaâ€™s inflation rate fell in February to its lowest level in more than three and a half years it was still running at an annual rate of 3.4%. We need to see average price declines in the region of 10% in both 2009 and 2010, and not only am I not convinced we are going to see that, none of the major bank analysts or multilateral organisations are currently forecasting anything like this. Or are we going to run our correction from now till 2015 (and have something which looks more like an L-shaped correction)?
What we really need to think about here is not the general index, however, but the so called “core” index (the one that excludes volatile items like energy, food, alchohol and tobacco). Now as we can see in the chart below this index has stabilised, and has even started falling slightly, but if we keep in mind the rule of thumb idea of a 20% decline, and note that the core level peaked at 118.37 in December, then for the correction to have any hope of working we would need to be looking at a reading in the region of 95 come December 2010.
And the situation may be even more complicated than we imagine, since the Eurozone itself may fall into deflation, and if so every percentage point drop in the Eurozone index will need to be matched by an extra percentage point drop in the Estonian one. Unfortunately your leaders and advisers are a long way from explaining this harsh reality to you.
But there is reason to fear that this may actually be what happens, since if we look at Eurozone headline HICP inflation on an annualised basis, we will find that it fell more than expected in January – to 1.1 per cent, according to Eurostat data – down quite dramatically from the peak of 2.7 per cent hit in March last year. This was the lowest level we have seen since July 1999, and a sharp drop from the 1.6 percent rate registered in December. On a month-to-month basis, prices were down 0.8 percent. The “core” inflation rate – that is consumer inflation without the volatile elements of food, energy, alcohol and tobacco – we find it still stood at 1.6%, since the biggest impact on headline inflation comes from the decline in food and energy costs. But if we look at the monthly movement in the core index, we find that it dropped by a very large 1.3% (see chart below).
Now if we come to look at the core inflation rate over the last six months, we find that the index has only risen 0.1% (or an annual rate of 0.2%). This gives us a much more accurate reading on where inflation actually is at this point in time, and where it is headed. The chart below shows the six month lagged annualised rate for the last twelve months, and the sharp drop in January is evident. If things continue like this, then the eurozone as a whole is headed straight into deflation, for sure.
Retail Sales Dropping Sharply
Basically, to get economic growth, and thus to be able to pay down debts, you need one of three things: an increase in government demand, and increase in export demand, or an increase in private domestic demand. Now the first two of these are categorically excluded in the present situation (especially since the government is cutting, and not increasing, public spending as part of the crisis response package (the so called “plan A” strategy). However, private domestic demand is falling like a stone at the moment. According to the latest data from Statistics Estonia, retail sales were down 10% year on year in January (at constant prices).
As we can see in the chart below, Estonian retail sales peaked in February 2008, since which time they have been steadily falling.
So what are the chances that domestic demand can make a recovery? Well, according to some, quite substantial. According to a recent report from UBS bank on Eastern Europe Lending:
We retain our firm view that convergence is a â€˜sure thingâ€™ for those economies already in the EU â€“ it is just a question of time before levels of GDP per capital approach those of the established members. If convergence is perhaps a thirty or forty year process, the most advanced are perhaps half way through (Poland introduced its free market reforms on 1 January 1990). The uncomfortable period we are entering is one where local growth goes from above-trend to sharply below. It may well take a number of years before nominal GDP (in Euro) recovers the levels of summer 2008, but we believe markets can be forward-looking when outcomes are predictable.
So the issue is convergence, and the justification for “plan A” is essentially based on this idea, as UBS analysts
Why does convergence matter so much? Because equity markets â€“ and therefore companies â€“ are essentially about growth. And convergence drives excess growth. The new EU members offer legal systems becoming increasingly like those in old EU states, with labour productivity comparable and labour costs a fraction of those back home â€“ particularly following recent currency declines. Margins on banking products are typically higher than in â€˜oldâ€™ Europe and levels of penetration much lower.
These arguments were a staple of a thousand corporate presentations through the good times and we suspect will be little mentioned except where necessary over the next twelve or eighteen months. But we believe them to remain essential to an understanding of likely outcomes in the region: they raise the bar for all stakeholders faced with a challenge of whether to prioritise the long-term or the immediate. It is an active debate what the Ukraine will look like several years hence; we believe it is not for the EU members: they will look more like the old EU states, in form and substance.
So we are putting all our money on the “convergence” bet, but just how realistic is this? Unfortunately, not very, since one key argument it simply fails to take into account is the effect of demographic processes. Basically, the whole of Eastern Europe has one large and little discussed problem, birth rates fell dramatically, but life expectancy did not rise: Latvia and Estonia are not only (along with Slovakia) the EU countries with the lowest per capita income, they are also those with the lowest life expectancy. Male life expectancy in Estonia is just 67.16, and for Latvia it is 66.68, compared to 76.11 for Germany, and 77.13 for Italy. Let’s not beat about the bush here, this means that each adult working male can contribute roughly ten years work less to paying down the country’s debts, and of course, extending the working age to 70 (25% of the Japanese population still work at 75) impossible. This is why the whole idea of “convergence” is a non-starter. And again, you don’t need to be an economics PhD from MIT to see this.
In the real world Estonia’s population is currently shrinking, which, with fertility around the 1.4 Tfr range is hardly surprising.
The birthrate has been rising (slightlly) in recent years, but as Afoe’s Doug Muir explains in this post here, this is more than likely going to unwind during the recession.
Interesting Fact #1: birthrates tend to drop during recessions, and the drop tends to correlate with both the severity of the recession and the speed of its onset. The current recession is looking to be a bad one, and it happened pretty quickly, so we can reasonably expect a sharp drop in birth rates. Makes sense, right? Babies are expensive; more to the point, babies limit your options. They make it harder to move to a different city, change careers, stop working for a while. When times are hard and uncertain, babies become a luxury. For individuals and families, a recession is a good time to put childbearing on hold.
Interesting Fact #2: all across Communist Eastern Europe, birth rates declined slowly through the 1970s and â€™80sâ€¦ and then crashed after 1990, dropping to very low levels and staying there through most of the decade. In some countries they bounced back a bit, in others not, but in almost all cases thereâ€™s a big â€œbirth gapâ€ from about 1991 until at least 1997, and often later.
Put these two facts together, and thereâ€™s a problem.
Indeed Statistics Latvia have already reported a 25% year-on-year drop in births in January 2009 (from 2310 in Jan 2008 to 1860 in Jan 2009), and looking at the Estonian Statistics we find that in January 2008 there were 1493 births and in January 2009 there were 1232. Again about a 20% drop year on year. Of course, one month’s data don’t prove anything, but since, as Doug points out, this is what the theory predicts, we should all be taking it seriously, and it should be taken into consideration when we talk about which kind of “correction” we want. It is no good saving the stream of external funding coming into your banks if you “meltdown” your population as you do it.
Unfortunately I haven’t noticed one single European leader who is seeing fit to even mention this issue – or the other, pending, one that when the recovery does come, if the Baltic countries are still stuck struggling with their pegs, the additional haemorrage out will be in young people looking for money to send home to their ageing and impoverished relatives, thus giving the whole demographic thing another turn of the screw.
The future already looks bleak enough in human capital terms, as this recent report from Statistics Estonia makes evident:
According to the Statistics Estonia, at the beginning of academic year 2008/2009, 154,481 pupils were acquiring general education, 27,239 vocational education and 68,399 students were acquiring higher education. The decrease in the total number of pupils is influenced by the number of pupils acquiring general education, which has decreased during the last decade. The decrease in the number of pupils in general education is related to the decrease in the number of births, which began at the end of the 80s and lasted till the end of the 90s. At the end of the 90s more than 220,000 pupils were acquiring general education, thus the number of pupils in general education has decreased by about a third during the last decade. In academic year 2008/2009, 147,519 full-time and 6,962 part-time pupils were acquiring general education. In autumn 2008, 12,426 children started school, which is over a third less than ten years ago.
So Is There A “Plan B”?
Well, of course there is, and everyone, no matter which side of the argument they are on, knows only too well what this is: devaluation. Of course of devaluation of the Baltic/Latvian pegs contains implied sovereign liabilities, and these need to be thought about. You cannoy do this alone, but you are members of the EU and you can ask for help with the process. But if you don’t start to ask for the help, then naturally you aren’t going to get it.
Technically the pegs can be maintained. The question which faces Estonians is quite simply which alternative â€“ keeping or changing the peg â€“ implies the greatest cost. The main stakeholder here is the EU, and you should be leveraging that for all you are worth. The capital erosion for Western European lenders would not be insignificant if you (and others) simply sink.
Naturally small open European economies like Latvia and Estonia can only hope to gain very minimal monetary autonomy outside currency board type arrangements, so the only realistic exit strategy is devaluation and Eurozone membership, as I explain in this post (and this one).
Of course this change in EU policy won’t arrive tomorrow (but it might come next week, or the week after). It’s just that you have to push for it. Stopping work and going home (as unemployed) while your country borrows more and more money is not going to bring the future you all so badly want. There is another path, choose it!
Here is an extra chart showing Estonian unemployment rates from 1993 to date, as provided by TÃ¶Ã¶turuamet, the Estonian Labour Market Board.
and here’s a longer time series for the Eurostat labour survey data. The difference between the two is methodological, and the Eurostat data is more comparable with other EU countries. As we can see, on both measures unemployment has been rising very rapidly of late, and with a 10% contraction forecast for this year, the end result is almost certain to go higher than the 13.2% peak registered in 2000 according to the Eurostat data.