Facebook Blogging

Edward Hugh has a lively and enjoyable Facebook community where he publishes frequent breaking news economics links and short updates. If you would like to receive these updates on a regular basis and join the debate please invite Edward as a friend by clicking the Facebook link at the top of the right sidebar.

Monday, October 29, 2007

Time To Cry For Argentina?

Two issues - apart from the exceptionally high levels of recent economic growth - have dominated the economic landscape which formed the backdrop to Argentina's latest election: the price of tomatoes and the availability of energy.

In many ways this is somehow typical of Argentine social and political life, the devil is always, or nearly always in the details, and the devil, in one form or another, is never, it seems, far away. As outgoing President Néstor Kirchner is quoted as saying at one point "It's not easy work helping Argentina find it's way out of hell, but in these past four years we've shown that it's absolutely possible to do so". So the moral is, it seems, that the devil makes hard work of life, but doesn't always win, unlike those who "adjust" the inflation numbers, who see to do so eternally.

In fact the really big, big news which lies behind the recent electoral triumph of Cristina Fernandez isn't the technique in pruning inflation data, it is the fact that Argentina's economy has been expanding at a sustained annual pace of 5 per annum over the last four years, and is, if anything, accelerating at the present time, since the Argentine economy reportedly expanded at the fastest pace in three months in August (growing at annual rate of 9.2%), leading it onward and upwards toward a fifth straight year of 8% plusgrowth.

Of course the big issue here is just how sustainable all this is. Is the Argentine economy, as some would have it, on the verge of yet another of those hyper-inflationary spirals, or is there something different about Argentina's evolution, this time round?

Certainly reasons why Argentina's "growth miracle" should not continue are hardly hard to come by. The first of these is the incipient rising inflation, and the government shenanigans in trying to hide it. The issue of food prices hit the headlines in grand style earlier this month when Argentine consumer groups launched a week-long boycott of tomatoes after prices reached almost $6 a kilo, making the salad vegatable which garnishes the nation’s dinner plates effectively more expensive than the meat it normally accompanies. The boycott itself took place amidst growing concerns about galloping food inflation and widespread criticism that the official inflation figures presented by the government were being manipulated as part of the run-up to this weeks elections.

Food prices are now widely estimated to be rising at an annual rate of around 22.5 percent - or more than double the inflation rate which is to found in other sectors like clothing, or domestic appliances. The "say no to the tomato” campaigners handed shoppers a tomato in a bag marked "officially priced at 3.99 pesos ($1.30), don’t overpay”, an ironic reference to the official price recorded by Argentina’s much-questioned National Statistics Institute (Indec). Incumbent President Néstor Kirchner has defended Indec’s figures, most notably by describing them as "perfect”, but after months of apparent government manipulation of official data there are few who are still willing to swallow this. Independent economists expect inflation to end the year somewhere in the 15 to 20 percent region, or at double the rate which is to be found in the official government’s forecast. At the present time inflation is estimated by Goldman Sachs to be about double the official 8.6 percent rate, and the blame is put firmly on a 35 percent jump in government spending this year.

To back their inflation claims Kirchner's critics have been pointing to a breakdown in the traditional relationship between price rises in Buenos Aires, the province that the government uses to gauge nationwide consumer prices, and prices in Mendoza province. In the 10 years up to 2006, annual inflation in Mendoza, a wine-producing region located in the foothills of the Andes, was on average 0.4 percent higher than in Buenos Aires. That gap however swelled to 9.2 percentage in the first seven months of 2007.

Indeed in August Patricia Gimenez, the head of statistics for Mendoza province, initially reported inflation as running at 3.1%, more than double the official 1.5 per cent month-on-month rate, but the figures were later modified by the National Statistics Institute to show the lower rate 1.5% rate, according to the newspaper Clarin, and the earlier figure was attributed to an administrative error.

As can be seen below, according to the government prepared Indec data, inflation would actually seem to have been slowing in Argentina so far this year.

Argentine Factories Running on Empty?

The second issue which has been bobbing steadily up and down in the background, and which has given lots of fuel to the sceptics, is the state of Argentina's infrastructure, and in particular the preparedness of its energy sector. Last July, and for the first time, even Néstor Kirchner himself was forced to concede that a major problem existed even to the extent of using the term “crisis” to describe the severe energy shortages that Argentinians had been experiencing, shortages which in fact forced the government to ration gas to thevery factories which have been fuelling all that recent GDP growth in an attempt to ensure that there was enough available for heating homes.

One of the threads which runs through Argentina's current energy woes is, of course, the issue of climatic change. Lower than traditional rainfall has increasingly been causeing problems for hydro-electric dams, and both Argentina and Chile have relied to some considerable extent on hydroelectric power to meet their energy needs.

There are, however, other issues, among them the relatively high level of energy interdependence among Latin American countries, and the relatively high level of political instability which characterises some of the countries concerned, most notably Bolivia and Venezuela. What all the interlocking and interdependence effectively means is that problems in one country are rapidly transmitted to another. At the end of June, for example, Argentina was only receiving 4.6 million cubic meters of the 7.7 million cubic meters of natural gas it needs daily from Bolivia.

Argentina is not the only country affected by this situation. Whenever Bolivia cuts exports because of internal chaos (which is a not infrequent event), or becuase of technical problems caused by an investment deficiency (which are ongoing), then it is contractually obliged to service its Brazilian export contracts before it services Argentina. The resulting shortages then prompt Argentina to reduce its exports of Bolivian natural gas to Chile. Chile also is facing natural gas shortages, but cannot purchase directly from Bolivia because of the ongoing feud between the two countries over access to the ocean, which Chile took from Bolivia in 1884. Chile will soon construct liquefied natural gas ports to overcome its energy problems.

Within both Chile and Argentina, residential consumption of natural gas (which in Argentina increased by a staggering 30 percent in May because of an early winter cold snap) is given priority over industrial use. This has led to shortages for power plants and factories in both countries, and also to the use of diesel and other petroleum-based fuels as a substitute for natural gas in those facilities capable of using multiple fuel types. That, in turn, has led to tremendous pollution problems, particularly in Santiago, and to diesel shortages that are affecting Argentine farmers during harvest (though the country's main crop, soy, has not been affected).

But is ever the case with Argentine economics, domestic political questions are never far away in the background, in this case in the form of Argentina’s ultra-cheap energy tariffs, which have helped fuel the increase in demand at the same time as they have helped the economy bounce back from virtual ruin in 2002. After four years of pretty spectacular GDP growth - averaging over 8 per cent per annum - Néstor Kirchner is now prepared to admits the dramatic economic growth has created "bottlenecks", but he has largely attempted to shift the blame onto energy companies for failing to invest enough to boost gas production, and transport companies for failing to deliver enough energy.

From a more conventional economic point of view, of course, the issue of the government regulated tarrifs lies behind the issue of the lack of investment. So while external factors undoubtedly played some part in the recent energy trauma which has shaken Argentina, to some considerable extent the problems have "made in Buenos Aires" written all over them. Since coming to office four years ago, President Nestor Kirchner has refused to lift price controls on utilities. These controls were originally imposed during the financial crisis of 2001-2002 to avoid social unrest, and afterward they were continued in order to give the appearance of controlling inflation. However, these fixed prices have discouraged new investment in the country's ample natural gas fields, leaving Argentina cripplingly dependent upon Bolivian imports, even as economic (and hence industrial) growth in the region exacerbates energy shortages. Although Argentina's natural gas production has increased over the past few years, it has not increased enough -- and most of the increases have not come from the opening of new fields, but from existing ones that are rapidly maturing and will soon "top out".

The Electoral Dynamics of Inflation Linked Bonds

On another front, the widespread suspicion that the government of Néstor Kirchner has been manipulating the inflation data and the likelihood that his wife Cristina Fernandez would succeed him have been steadily transforming the Argentine bond market. Argentina's benchmark inflation-linked bonds have fallen 24 percent in the course of this year, making the country's debt market the worst performer among any of the emerging economies, according to data compiled by JPMorgan Chase, and has converted Argentine inflation-linked debt into the single worst long-term asset to be found in any of the emerging-markets. Yields on Argentina's 5.83 percent inflation-linked peso bonds due December 2033 have soared to 7.7 percent from 5.3 percent on Feb. 1, which is just shortly after Kirchner began making the personnel changes in the statistics office.

About 40 percent of Argentina's $136 billion debt is in inflation-based securities, and the value of their their principal rises and falls along with the consumer price index. Merill Lynch estimates that Argentine bondholders may well have lost out on around $250 million in interest payments this year alone, and by steadily reducing the official rate the Argentine government stands to save something in the region of $5 billion in principal payments at maturity.

Of course in electoral terms tampering with the index for the price of tomatoes is one thing, and tampering with it to reduce the size of overseas debt quite another. Whilst the polls show that most Argentines feel that consumer price increases are accelerating, and this obviously is a matter of concern for people who have lived through inflation which skyrocketed up towards the 20,000 percent level in the early 1990s, there are also other items on the household balance sheet which help to some extent to explain Cristina Fernandez's popularity. Like unemployment, which has fallen steadily in recent years.

Surviving The Emerging Markets Storm?

In fact, despite the numerous tricky issues which arise in association with Argentina's sovereign bonds, the country itself and its financial markets have largely been spared the kind of problems which have been experienced in some of the other emerging markets like South Africa, Hungary or Turkey. One of the principal reasons for this is that now – and in stark contrast to the situation which characterised much of the 1990s – Argentina's economy is largely protected against external financial pressures by the presence of budget and current account surpluses, and with these a steady increase in foreign exchange reserves as well as a reduction in financing needs. In addition the rather poor reputation which Argentine national debt and institutional quality enjoy have meant that the peso, rather than appreciating against the dollar as many emerging currencies have found themselves doing, has in fact been trending slowly down. This has made it comparatively easy for the Argentine administration to maintain competitiveness and a trade surplus in the face of strong internal inflationary pressure.

In other word, the Argentina administration hasn't found it necessary to follow in the footsteps of some of the other emerging market countries and adopt a currency peg (something which in any event would be unthinkable in Argentina following the experience of the late 1990s), since market sentiment has done effectively done the work of weakening the peso for it. In fact the peso has fallen steadily against the dollar during the course of the last two years, at a rate which has steadily eaten up inflation but without causing any form of major financial disruption.

Moreover, Argentina’s real economy is much less vulnerable to external shocks than it used to be, and while it was affected by the market turmoil of last August, it has hardly been knocked off its orbit by it. Part of the explanation for this is that one of the consequences of the 2001 crisis has been that there have been few capital inflows into Argentina, and indeed there have been net capital outflows in recent years, a situation which stands in stark contrast with the high levels of foreign capital penetration which existed in the late 1990s. The consequence of this is, naturally, that the level of external debt has fallen steadily.

At the same time the level of public debt has been steadily reduced. Argentina's tax revenue was up 37.4 percent in August 2007 in comparison with August 2006, on the back of strongly rising wages and sustained consumer demand, and rose to 17.9 billion pesos ($5.7 billion) from 13 billion pesos a year earlier. Revenue from value-added taxes rose 43.3 percent to 1.8 billion pesos and income tax revenue grew 33.4 percent to 960 million pesos. So while government spending has increased significantly in this election year, the strong revenue inflows have enabled the administration to service the spending without increasing the level of public indebtedness.

And the boom in Argentina is hardly a credit driven one. Argentine banks only lend the equivalent of about 40 percent of their deposit base, a figure which is roughly half the average rate prevalent across Latin America. At the present time loans by banks equal about 11 percent of gross domestic product, and again this is down from the level of around 30 percent which prevailed before the financial crisis.

Indeed all of this recently lead Nestor Kirchner to chide Argentina's banks, who charge an average credit card interest rate which is roughly double that which is current in the United States, for example, urging them to cut finance charges and lend more, even vaguely threatening that if they didn't act, then the government would consider introducing rules to force them to do so.

``Banks are too liquid, have lots of money saved, I'm glad they are solvent,'' he said recently in a speech at the presidential palace. ``But you have to give loans at low rates, otherwise I'll have to take some measures.''

However, if we look at Argentine overnight interest rates, they are hardly high in comparison with those prevailing in some emerging markets given the level of inflation which prevails (even on the official version), and clearly it is totally unrealistic to imagine that the banking sector can offer rates of interest which fail to cover them for inflation.

But while the central bank administered overnight rates are far from high, annual credit card interest rates, in comparison, rosen to 26.85 percent on average in July, up from 25.93 percent last December. Rates for personal loans rose to 24.74 percent from 24.56 percent over the same period. By way of comparison we could note that the average fixed rate for credit cards in the United States is currently in the region of 13.5 percent.

So if Argentina is badly governed, and infrastructurally and institutionally is far from being a shining example to others, what exactly is going on. Why should Argentina have done so badly at the end of the 1990s, when it was being a model student, and why should it be doing so notably well today, when on many counts it is doing exactly the opposite of what the textbook recommends.

It's the Demography Silly!

Well undoubtedly there are many reasons for this, and China's rpaid growth and demand for Argentina's agricultural products would be high on the list, but if I could single out one factor which I think stands out above all the others it would come in a single word: demographics. So let's take a quick look at why.

First off, and as is evident, Argentina's fertility rate has been falling for many years, and is now about to go below replacement level.

This fall in the fertility rate has lead to a steady easing off in the rate of increase in the number of children born, and since 2000 the number has levelled off, and even begun to decline slightly.

At the same time life expectancy has been improving, and most notably life expectancy among the young and poor, where we can note the steady decline in infant mortality which has again been taking place.

As a result of these processes Argentina's population is still increasing, but again at a much slower rate than hitherto.

One consequence of all of this is that the percentage of the population in the 0 to 14 age group has been declining steadily from the high point reached at the end of the 1980s.

All of this is producing profound changes in Argentina's age structure, changes which are associated with what is known in an economic context as the demographic dividend. These hanges can be seen reasonably clearly in the following population pyramides for 1990, 2000 and 2010. We can see how the shape of the pyramid begins to change, and particularly in the third pyramid how the steady stabilisation, and then decrease, in the number of children born means that the generation size starts to shrink. It is the appearance of this change at the base of the pyramid which is the most typical indicator of the presence of the demographic dividend.

One of the reasons why this moment is so favourably in economic terms is the impact it has on saving and investment. Basically the relative decline in the proportions of young children frees off a greater proportion of national resources for saving and investment, and it is this process which means that the dependence on external funding begins to decline. A similar process can be observed in China, and is now being observed in India.

So to end where I started, with the question - should we be crying for Argentina? My answer would be most definitely no. There may be a lot, a hell of a lot, wrong with the way Argentina is being run, and some of the issues have been adressed in this post. But the underlying situation in Argentina is far from being a tragic one this time round. Clearly Argentina needs to move away from these 'old practices' of manipulating the statistical system. Clearly Argentina needs to attract Foreging Direct Investment in order to modernise the energy sector and its infrastructure generally. But the situation is far from being a hopeless one, and reform is not only possible but probable. Not only that, more by accident than design Argentina may go into the next global growth slowdown better equipped to be able to withstand the pressure than many others. If so this will also be a test, a test for the relative importance of institutional quality when measured against the driving force of demographic tailwinds. The outcome promises of this testto be interesting, very interesting for all of us.

Saturday, October 13, 2007

Is An Emerging Market Correction Coming in Eastern Europe?

According to wikipedia, the noun "correction" comes to us, via the French, from a Latin original corrigěre 'to make straight (again)', and is used to describe an action which can rectify, or make right a wrong. Wikipedia lists a number of possible meanings, three of which are of interest to us here:

  • To set straight an error, clarify a misunderstanding, undo resulting damage; e.g. a correction in a newspaper is the posting of a rectification of a mistake that appeared in a previous issue of the newspaper.
  • To rectify an abnormal state of affairs - e.g. a market anomaly - as occurs, for example, in a stock market correction.
  • A euphemism for a punishment, which may be of various kinds, mainly physical; in institutional terminology specifically used for imprisonment, e.g. correctional facility (prison) or corrections.

Basically wikipedia give us a very reasonable account of the useage of the term "correction", and of its ambiguity in an economic context. "Correction" it seems means both punishment and putting right. What is so unique and so surprising about the coming correction in the East European economies is that it fits the definition of correction in neither of these senses.

In the first, and most obvious case, if there is to be a punishment, then there must have been a crime, right? But where is the crime in the current case of Eastern Europe - unless it is to be some modern variant of original sin - since it seems evident that the only things that policy makers in the EU10 economies have been doing is "following orders", irrespective of whether these came from the EU Commission, the ECB, the IMF or the World Bank. They took the best expert advice available at the time, and they acted on it. Surely they can hardly be blamed - let alone punished - for that. If there have been errors, they are surely human ones, a bit too much public spending here, the odd currency peg or other there, but when we look beyond all this surface detail, and move to the underlying structural level, we find that there is a most depressing uniformity about things. I say depressing, since I do think it is hard to say the Eastern Europe has done much to bring about or deserve the tragedy which is about to unfold and come crashing down upon its head.

If the twentieth century was - for no fault of their own - a bad one for many of these countries, it now seems that the twenty first one - oh horror of horrors - might turn out to be even worse.

Why do I say this? Because many of these countries are quite literally dying, in the demographic sense. And since at the present moment almost all of them - with the honourable exception of Hungary - are working pretty much flat out, it might not be going to far to suggest that they are "dying on their feet" or "dead men running (on empty)" even.

Which brings us back to the second meaning of correction, that of putting straight or "making whole".

Unfortuantely this is just what the coming correction is not going to do. The explanation for why it will not do this will come in the following post, but suffice it to say here that - due to major underlying demographic processes and the impact of migratory flows - the traditional homeostatic mechanisms which operate during economic corrections will not be at work in this one.

That is to say, what is about to happen next in Eastern Europe is about to establish, and I suspect beyond all possible doubt for the company of reasonable men, that the widely accepted neoclassical idea of general economic convergence towards a - even hypothetical - situation of "steady state growth" is quite simply a mistaken and non-valid one. Demography does matter, and fertility with a lag of twenty or so years does seem to be important.

Now lets take a look at why.

The Problem-Set Facing the EU10

Basically the principal outstanding issues confronting the EU10 countries are threefold:

1/. Labour capacity constraints (which are normally a by product of long-term low fertility and large scale recent migration flows) are producing significant wage inflation and strong overheating.

2/. A structural dependence on external financing - which is in part a by-product of the effect of low levels of internal saving, and which is another factor which separates the EU 10 from those like India or China who are benefiting from a typical demographic dividend driven catch up, is leading to large current account deficits, and potentially high levels of financial instability.

3/. A loss of control over domestic monetary policy due to eurozone convergence processes which - with or without the presence of formal pegs - make gradual downward adjustment in currency values as a alternative to strong wage deflation virtually impossible. This issue is compounded by the likely private "balance sheet consequences" of any sustained downward movement in the domestic currency given the widespread use of mortgages which are not denominated in the local currency.

Now the worrying part about all three of these is that they are not simply cyclical in character. As such they are not problems which will "self correct" as a result of a recessionary slowdown, whether this be of the "soft-" or "hard-landing" variety. This problem simply is not being taken into account in many of the current pronouncements on the EU10, and certainly is in no way reflected in the current "fiscal deficit obsession" which we can find in the discourse which exists at EU Commission level (for a consideration of this in the case of the Czech Republic see this post).

The IMF, the World Bank and the ECB

The macro imbalances which currently exist in many of the EU10 economies are substantial, and as I have tried to argue at some length here, they stem from a virtually unique set of circumstances (historically unique I mean, at the present time the underlying dynamic across all the East European EU member states is remarkably similar, with the possible exceptions of Hungary and Slovenia, and in each of these latter two cases for different reasons).

The sad reality is that many of the EU 10 countries (and in particular we are talking here of the Baltic States, Bulgaria, Romania and Poland) are facing, at one and the same time, a very considerable inflow of external funds to fuel domestic consumption (whether this be in the form of the bank flows which drive the supply of cheap credit, or the worker remittances which drive the demand for it), and a very significant reduction in the potential labour supply following a couple of decades of below replacement fertility, and a large and sustained outflow of migrant workers which has accompanied EU accession and which is driven by the very large East-West wage differentials. Put another way, demand side factors are increasing rapidly, while supply side capacity ones not only are unable to keep pace, but are actually shrinking (if we think about the number of people of working age in the total population, and of the proportion of these who are available for work inside the country).

In recent weeks a variety of players on the international economic stage have been active in pointing to the mounting problems which are now only to obvious to the trained macro economic eye.

In the first place the World Bank, in its most recent report on the economic outlook for the EU 10, had the following to say about the labour shortage situation:

Unemployment has fallen substantially in virtually all EU8+2 countries since 2004 due to strong growth in labor demand. This has given rise to skill shortages and associated wage pressures, often amplified by out-migration of EU8+2 workers.....

In contrast to the earlier period of weak labor demand it is now the supply side of the labor market that constrains new job creation. Many persons of working age are economically inactive in EU8+2 either because they lack skills demanded by employers, or because of labor supply disincentives, such as early retirement benefits, generous disability schemes, high payroll taxes, and limited opportunities for flexible work arrangements. These effects are concentrated among the younger and older workers, while the participation rates for middle aged workers are similar to those of the EU15. Hence the main challenge facing now EU8+2 is to mobilize labor supply to meet the demand.

Really anyone seriously interested in this problem needs to read the whole report, but the above, in essence, is one part of the picture.

Then, earlier last week, the IMF published the autumn edition of its World Economic Outlook, and Chapter 3 - Managing Large Capital Flows - to some extent focuses on the current problems of Eastern Europe. According to the IMF the economies of eastern Europe are vulnerable to a reversal in the surge of private capital that has poured into emerging markets in recent years. The IMF says that "large capital inflows are of particular concern to countries with substantial current account deficits, such as many in emerging Europe", as well as countries with inflexible exchange rate regimes (and among these are, of course, to be found four EU 10 members: Estonia, Latvia, Lithuania and Bulgaria).

Eastern Europe is in fact the only region singled out by the IMF in this way as being particularly vulnerable to a change in the direction of private capital flows at this point in time. While the IMF draws the fairly comforting conclusion that most emerging markets now have much stronger and more defensible current account positions than they did in the late 1990s - since they have been steadily building up their foreign exchange reserves during the good times - this is not the case with most of Eastern Europe. Gross capital inflows into the region have reached levels relative to gross domestic product that are "unprecedented for emerging market countries in recent history" while "unlike in other regions…net capital inflows have been accompanied by a deteriorating external position".

So, I would say, we have been warned. It couldn't get much clearer than this.

Lastly there have been a number of recent indications that over at the ECB people are not all that calm about what is happening at the moment. The most recent example of this unease is to be found in a speech given by European Central Bank board member Lorenzo Bini Smaghi where he specifically identified the fixed exchange rates in Bulgaria and the three Baltic states as part of the cause of the accelerating inflation and current-account deficit problem which exists in those countries since the currency board arrangement limits the central banks' ability to control the rate of price and credit growth. According to Bini Smaghi:

Keeping nominal convergence on track is the main policy challenge in the region. The problem becomes particularly acute in countries which have given up monetary policy independence by choosing an exchange rate target or adopting a currency board arrangement.

The key question for these countries is: how is it possible to keep inflation under control by pegging the exchange rate, which means adopting de facto the monetary policy of the euro area, especially since the euro area economy is growing at a rate that is less than a third of what a catching-up economy should aim to achieve? In other words, how is it possible to keep inflation under control with very low or even at times negative real interest rates? What are the risks for financial stability of having persistently low real interest rates, much lower than the rate of growth of the economy?......

Bini Smaghi has it right, the key question for the EU 10 countries is how to maintain the levels of "catch up" growth which would enable them to close the gap in living standards which exists between East and West, and how to do it, so to say, when they don't have the raw material (in terms of labour supply) to hand to aid them in this.

So, at the risk of repetition, there are three large and oustanding problems to be faced in Eastern Europe now.

1/ The labour shortages issue

2/ The capital flows, current account deficit, dependence on foreign investment, non local currency denominated debt issue

3/ The currency peg issue, in the case of the four countries which are on a peg, or the rising real exchange rate one in some of the countries which aren't like Romania, Poland and Hungary. Given the dependence on foreign lending for survival, the difference at the end of the day is not significant between these countries, with the possible exception of the fact that any attempt to break the peg by one of the four countries who have one could prove to be a very dramatic event, and indeed could serve as the immediate signal for setting off the whole large scale "correction" which is so evidently in the process of coming. It's in the post, as it were.

In this situation the only real tool left to the domestic authorities is the generation of fiscal surpluses to try and reduce domestic demand, to try, as it were to dredge domestic demand from the system. But if this is not to remain something akin to draining the ocean with a soupspoon these surpluses need to be very large indeed. The IMF has been arguing for a surplus of 4% of GDP in the Latvian case, and it is far from clear that this in itself would have been sufficient (I say would have been, since, of course, the closing down of the credit mechanism which is being already operated by the banking system may well now make this unnecessary, and this closing down is happening very rapidly indeed, as Latvian abroad makes plain here). Naturally the local authorities - not really grasping the gravity of the situation, and who can blame them since who really anticipated the severity of this problem - have flinched in the face of introducing such severely deflationary fiscal surpluses. As Bini Smaghi says:

"I guess the real question to ask is: how large should the budget surplus be to counteract the inflationary effects produced by a pro-cyclical monetary policy and would this be acceptable for a catching-up country? How far reaching, and acceptable to the population, should structural reforms be? All in all, the requirements for the budgetary and structural policies associated with an exchange rate linked to the euro might just be too demanding to counteract the procyclical effects of very low real interest rates. This might lead to boom and bust cycles, with potentially very severe adjustments costs that may delay real convergence."

I entirely agree, and what we are faced with here are a whole new set of problems for monetary policy, ones which were not at all anticipated when the euro was set up, or the new members accepted into the EU fold.

Why Doesn't Homeostasis Operate?

And so to finish where we started. This correction will not correct, at least it will only correct in the punitive sense, since little that is good can come out of the affair. To understand why this is you need to look at the population pyramids of the various countries involved.

To help make this a bit plainer I have comparative pyramids for Ireland and Latvia online here. This comparison basically is to illustrate why a country like Ireland was able to get a demographic dividend which facilitated so much rapid catch-up growth, and why Latvia isn't.

Now if we look at the pyramid for Turkey in 2000:

and if we now compare this with pyramid of Latvia's population in 2000, we should be able to see straight away why Turkey could correct rapidly, and why Latvia will not be able to.

Essentially if we look at the size of the "10 to 14" and "15 to 19" generations in comparison with the "20 to 24" one in both cases. Think of water running down a river, and the recession/correction as a kind of dam you place across the flow. In the Turkish case closing the dam gates causes the water to back up in the system and accumulate, ready to be subsequently released as part of a flood which due to its volume maintains wages at a very low level, which is exactly what we have just seen take place in Turkey post 2000.

If we look at a chart of Turkish GDP growth from the late 1990s we can see this process at work very clearly:

During the years 2000 to 2002 the Turkish economy underwent a correction in the true sense of the word. There was a problem, and it was corrected. The evidence for the effectiveness of this correction can be seen in the subsequent growth rates, and even, as I explain in depth in this post, from the way in which the Turkish economy has rebounded from the currency crisis of the summer of 2006. The Turkish economy is now reasonably robust, and one of the reasons that it is so robust - given that it has made more or less similar structural reforms to all the economies of Eastern Europe - is that it has the demographic winds behind it, and not, as in the EU 10 case, blowing into its face.

In the Latvian case, on the other hand, (and for Latvia here, of course, read any one of the other EU 10 countries which you please) stopping the flow temporarily doesn't correct the problem since in each time period there are progressively less people arriving to the gates of the labour market dam. It could even be that the correction here may only make matters worse, since it could in fact send many more potential workers out of the country in search of work. So far from bringing wages back into line with a competitive level, wages could in fact get "stuck" and only adjust downwards slowly, far too slowly to generate the new employment which will be needed to seriously restart the economy.

This is also why all those comparisons between what is happening in the Baltics now and what happened in Scandinavia in the 1990s miss the central point, since the demographic situations of the countires involved was very different.

Of course the EU10 still have the potential labour force left to get economic growth, what is in question is where they are going to get the labour supply from for the very rapid rates of catch up growth they obviosuly want and need.

And all of this has been a very long and tortuous way of saying that (and to answer my own question) yes, I think a correction is coming, indeed it is already arriving, and it will be a serious one whose impact will be felt right across the eurozone. But at the end of the day it may well not do what corrections by their very nature are supposed to do, namely correct.


By the very nature of the limitations of a blog post, a lot of assertions have been made above which are, to some extent, in need of substantiation. Below I list a number of previous posts where some of the issues raised are treated in more detail.

Firstly, the issue of neo-classical growth, and why this is important. A good starting point on this topic would be this post here.

It has been asserted that the EU 10 economies are overheating due to labour shortage problems. This is explained in the Latvian case in great depth here.

Claus Vistesen does a similar job for Poland here, and for Lithuania here.

I have examined Romania in depth here, and Bulgaria here.

As has been argued, in fact not all the EU 10 economies are by any means identical, and the problems do vary. Despite sharing a similar demography with the rest the cases of the Czech Republic and Slovenia seem to be very different. One of the factors at work here is undoubtedly the existence of migratory flows within the EU 10 themselves, and in this sense not all will or need to follow the same trajectory (although of course at the end of the day the numbers simply don't add up, so there will be losers as well as winners). I have made some initial attempt to look at the situation of the Czech Republic here.

Hungary is again a very special case, and is certainly not overheating at the present time. I have a whole blog dedicated to trying to understand what is happening in Hungary and why it is so different, this post and this one might be a good place to start.

One common assumption is that labour shortages in the EU 10 can be met over the longer term by inward migration from other parts of the old Eastern Bloc like the current CIS countries. Immigration may well be part of the answer for the EU 10, but the CIS itself is unlikely to be a durable source of supply here, basically because a large part of the CIS has a similar demographic profile. This post goes into the Ukraine situation in some depth, and here is a link to some work by Russian economists who advise the World Bank, and who argue that Russia may well be already in need of around a million migrants a year if she herself is to be able to maintain current growth rates.

I even went so far as to check out Uzbekistan, since many seem to think that these Asian members of the CIS might be better placed, but unfortunately my conclusion was that with all the demands which are being placed on them, and all the money which is heading back home in remittances to fuel growth it won't be long before they run out too.

Finally, if you have gotten this far, all I can say is thanks for your time, and good luck and good reading.

Friday, October 12, 2007

The Implications Of Bulgaria's Inflation Surge

Bulgarian inflation, which is now rising faster than in any other country in the European Union, accelerated once more in September on the back of rising food, wages, utility and education costs. The Bulgarian "major groups" index rose to a nine-year high in registering an annual rate of 13.1 percent, a figure which was up from the 12 percent one registered in August. Consumer prices rose 1.3 percent on the month following a 3.1 percent increase in August.

At the same time Bulgaria's annual inflation rate, as measured by the EU- Harmonized Index of Consumer Prices, rose to 11 percent in September, from 9.3 percent in August. On this index monthly prices climbed 1.2 percent in September, after rising 2.2 percent in August.

Now one common explanation for this inflation acceleration phenomenon has been the impact that the hot and dry weather at the end of June and in July - which was then followed by floods which also destroyed crops - has had on food prices prices. But there is obviously more to this situation than food costs. In the chart above I have included both the Bulgarian "home grown" index and the EU designated harmonised one, and, if we look at the latter, we can observe the way the two lines diverge back in May just as the food price issue started to gain traction, so we can see the food effect there in the divergence. The EU harmonised index evidently gives a lower overall weighting to food.

Food prices in fact account for 35 percent of the Bulgarian consumer-price basket, and food rose 2.1 percent in September, following a 7.3 percent increase in August. This gives us a whopping 25.1% year on year food inflation rate in September. Other factors which played their part like restaurant prices, which went up 17% year on year.

But going back to the indexes, we can see that even on the harmonised basis, Bulgarian prices are "only" rising at an annual rate of 11%. But this precisely gives us a measure of the extent of the problem. It is clear that inflation has suddenly accelerated in Bulgaria, and given the way the Lev is pegged to the euro, it is hard to see what the central bank can do, since raising interest rates would, in the short term, only attract a greater inflow of funds, accelerate the overheating even further, and put strong upward pressure on the currency.

Whichever way you look at it all of this is far from over, and inflation looks set to continue at a high level, if indeed it doesn't continue to accelerate. The International Monetary Fund recently doubled its Bulgarian inflation forecast for this year to 8 percent. This now seems like it might be a conservative estimate.

Beside food one factor is certainly of growing importance: wages. And wages are, as we all know, one way or another associated with the supply of labour. And it is just in this department that countries like Bulgaria find themselves in difficulty, as the world bank recently highlighted in a report on the growing labour shortages the EU 10 countries now face. Getting a clear fix on the labour supply situation in Bulgaria is something of a headache, however, given the absence of dependable data from the Bulgarian government on Bulgarians working abroad.

This issue is an important one, since if the Bulgarians are not in Bulgaria then they are evidently not able and willing to work at the time and in the place that the employment market needs them. On the other hand if we look at the evolution of the official numbers of unemployed in Bulgaria, we will see that they have been reducing quite substantially over the last two or three years.

If we then look at the data for the number of people employed, we will see that this has been steadily rising:

Now obviously these two graphs simply don't match, since they are self evidently set on collision course, and especially in a society which has now been having very low fertility for nearly twenty years and which has a significant proportion of its working age population outside the country working elsewhere. So if we look at the next chart we can see the quarterly year on year increases in those employed. That is to say we can see how many extra people were employed in the previous year.

As we can see the "assimilation rate" of new workers into employment has been accelerating since the end of last year, and indeed in the second quarter there were 312,000 extra workers employed, a feat which would be quite impossible over the next year (if the same participation rates are maintained) since there are currently only 237,000 registered unemployed in Bulgaria, and of course the problem will not be resolved by younger labour market entrants joining the labour force, since from now on pregessively more people will be leaving than will be joining, so this factor only makes things worse.

Naturally with the labour supply being exhausted at this rate, a response from the wage rate was only to be expected, and we got it.

As we can see, wage rates have been increasing fast since the start of 2004 (and have risen by nearly 50% over that time) and have been accelerating rapidly as this year has progressed. So this story is no longer only about food. It is in large measure a labour supply capacity problem.

As such, the issue is not a "homeostatic" one, in the sense that a recession will not bring things back to where they were before, since all that will happen as the years pass is that even less people will be available for work, due to natural population drift, even assuming the best case that another wave of people do not leave during the recssion. That is a "correction" in this case will not do what it is supposed to do - namely correct - since the Bulgarian population structure is now so badly distorted, it is hard to see what can actually be done at this stage. And the situation has all the hallmarks of being about to turn critical at just this very moment in time.

Of course, the position is complicated even further by the presence of the euro peg, and the currency board, since this to all intent and purpose leaves the central bank standing on the touchline as a virtually helpless spectator.

Naturally, the press has been full of gentle comments in recent weeks about how it might be adviseable for the Baltics and Bulgaria to begin to consider whether it might not be in their interest to steadily abandon the pegs.

Only last week European Central Bank board member Lorenzo Bini Smaghi gave a speech where he blamed fixed exchange rates in Bulgaria and the three Baltic states for accelerating inflation and current-account deficits because the currency board system limits the central banks' ability to control surging price growth. According to Bini Smaghi:

Keeping nominal convergence on track is the main policy challenge in the region. The problem becomes particularly acute in countries which have given up monetary policy independence by choosing an exchange rate target or adopting a currency board arrangement.

The key question for these countries is: how is it possible to keep inflation under control by pegging the exchange rate, which means adopting de facto the monetary policy of the euro area, especially since the euro area economy is growing at a rate that is less than a third of what a catching-up economy should aim to achieve? In other words, how is it possible to keep inflation under control with very low or even at times negative real interest rates? What are the risks for financial stability of having persistently low real interest rates, much lower than the rate of growth of the economy?......

I guess the real question to ask is: how large should the budget surplus be to counteract the inflationary effects produced by a pro-cyclical monetary policy and would this be acceptable for a catching-up country? How far reaching, and acceptable to the population, should structural reforms be? All in all, the requirements for the budgetary and structural policies associated with an exchange rate linked to the euro might just be too demanding to counteract the procyclical effects of very low real interest rates. This might lead to boom and bust cycles, with potentially very severe adjustments costs that may delay real convergence.

Bini Smaghi has it right, the key question for the EU 10 countries is how to maintain the levels of "catch up" growth which would enable them to close the gap in living standards which exists between East and West, and how to do it, so to say, when they don't have the raw material (in terms of labour supply) to hand to aid them in this.

Thus in many ways the European Central Bank might be thought to be increasingly giving the impression they would not be displeased if the Baltic nations and Bulgaria drop their exchange-rate pegs because they contribute to increasing economic imbalances, according to a research note from Danske Bank:

``It seems that the ECB is suggesting what would have been unthinkable a year ago: that it is time to change the exchange- rate policies in the CEE countries with exchange-rate pegs,''

This, at any rate is the view of Lars Christensen, a senior analyst at Danske Bank. No smoke without fire is normally a good watchword in these matters.

But the problem is that these countries will find it very difficult, if not nigh impossible, to break the pegs, quite simply because of the balance sheet consequences of all the foreign currency denominated (largely euro denominated) debt the citizens of these countries have been busily acquiring in recent years, and with the reasuurance that with a peg in place, and euro membership nothing could possibly go wrong. Except it can.

I think I have said enough here for one post, so I will not go into this issue further now, other than to point out that Claus Vistesen has conveniently done a timely post on the theoretical literature on balance sheet effects so we can all to some extent prepare ourselves for what might well now be about to happen. Claus will also be covering in more depth the capital flows and foreign debt side of things. So that will have to be it for this note. I doubt however that you have now heard the last of all this, or that much time will now pass without fresh news on this whole front, especially after the recent World Bank EU10 labour shortages report and the latest issue of the IMF World Outlook which focuses on the CA and debt imbalances in Eastern Europe. Basically the acceleration we have seen in inflation across a number of the EU 10 economies will have set alarm bells off ringing all over the place. The fire may well not yet have started, but I doubt it will be long now in kindling itself.

Thursday, October 11, 2007

Is Estonia Really Heading For A 'SoftLanding'?

Estonian inflation surprisingly accelerated in September to reach a new nine-year high, largely on the back of rises in food, housing and services costs. The Estonian Finance Ministry has said it expects Estonia's inflation rate to remain near the current level for the immediate future. The annual rate rose in September to 7.2 percent, which was its highest level since October 1998, and was up from 5.7 percent in August, according to data released by the Estonian statistics office at the end of last week. Month on month, prices rose by 1.1 percent.

The largest increases were registered in food prices (7.8% annual rate), housing (16.1%) and services (hotels, cafes and restaurants, 12%). Here is the monthly inflation chart.

Now the Estonian central bank and Finance Ministry have been insistent in recent months that the Estonian economy remains on course for a "soft landing" as the property sector cools and credit growth slows. This outcome is far being a self evident one, however, as this and other recent data which will be reviewed in this note make clear.

Overheating in the Baltics?

What is happening in Estonia may seem to be lacking in any great global significance, given the size of the country and its economy, and its lack of general importance at the global level. But such a conclusion might be over hasty, and premature, as I will try and argue here, given how representative what is happening in Estonia now is of processes at work in many EU10 economies (although not of course all, as the notable exception of Hungary shows).

In fact all three Baltic economies show clear signs of overheating, despite the fact that assessing the extent of the overheating in the Estonian case was initially complicated by an overly negative preliminary Q2 2007 GDP relase, which cited figures that were subsequently revised substantially upwards. I have addressed some of these questions in this post, and Claus had dealt with the external balance sheet position in this note here.

Slowing Down Nicely?

When we come to look at the details we can quickly appreciate that the Estonian economy certainly does show some clear signs of slowing, as can be seen from the GDP quarterly growth rates:

Also, and according to data from Statistics Estonia, industrial production increased in August 2007 when compared to August 2006 by only 4.4%, a pace which which was slightly down from the annual 5.2% increase recorded in July, and indeed the year on year rate of output growth has been slowing now since April, as can be seen from the chart.

But other data show if not a completely different picture, at least a more qualified one. If we look at retail sales for example, and again according to Statistics Estonia, we find that in August 2007 retail sales were up 13% year on year over August 2006. This rate is actually down on those registered earlier in the year, but it is still very high.

Actually, compared to July 2007, the level of retail sales in August was more or less stationary:

And again if we look at the monthly annual change figures, the rate of increase is clearly slowing.

Wage Push Inflation?

One big part of the underlying isssue here is undoubtedly strong wage-cost push inflation, of the old school kind. If we look at the quarterly chart we can see that though the rate of annual increase in wages and salaries has fallen back somewhat in the second quarter from the first, it is still very substantial:

Now if we move on to look at producer prices, which is where all the "cost-push" inflation shows up, we can see that these have been rising steadily over the last 18 months, and while they may well now have peaked in terms of the rate of increase, the pace is showing no real sign, up to now, of having eased off to any substantial extent.

The situation is even more problematic if we come to look at the rate of producer price increases in the export sector:

The reason this situation is so problematic is the impact that this sustained inflation on producer prices is bound to have on Estonia's ability to export, especially given that the kroon is effectively pegged to the euro, so prices cannot adjust via a downward movement in the currency.

We can perhaps get a point of comparison if we take a look at what has been happening in Hungary. The Hungarian economy has certainly been undergoing a "correction" since the autumn of last year, internal demand is down very sharply, and while domestic inflation remains stubbornly high - though not as high as Estonia's - export prices show a completely, and much more desirable pattern, since wage deflation has been very significant in Humgary.

Here is the chart for month-on-month changes in domestic sales prices and export prices in Hungary:

and below are the equivalent year on year changes. What we can observe is that there is now a very strong disinflationary process at work in Hungary, a disinflationary process which has yet to be seen in Estonia, or for that matter in the Baltics generally.

Now I think I need to be clear here, since I am certainly not recommending the kind of strong wage deflation they are having in Hungary as a preferred recipe for the Baltics. I am simply trying to suggest that this problem exists, and must eventually be addressed, either by coming off the peg, or by other means. There are even doubts in the Baltic acses that wage deflation as such could be operated, given that omnipresent danger of increased outward migration is bound to make wages and prices more "sticky" than in the Hungarian case (where among other things out-migration has not been present to any significant extent, at least to date it hasn't).

What I am suggesting is that there may well be other ways available to address the problem, even if they do involve "unconventional tools" and "out of the box" thinking. One of these unconventional measures would certainly be a flexibilising of the labour market through a significant and substantial opening to international migrant labour. In order to work this opening would have to be large scale, and should not simply be confined to skilled worker categories. At the end of the day, it is a question of which you prefer, to be flayed alive by systematic wage deflation (and all the problems of out migration that this might produce) or a rapid transition to a modern multi-cultural society. There are not many other options to play around with here I think. And time is pressing.

Now, returning to what I said above, the reason why all of this producer price inflation is so problematic for Estonia is the corrosive impact that it has on Estonia's ability to export its way out of difficulty. This impact is evident enough if we take a look at the recent evolution of the external balance in goods and services for Estonia:

Not a pretty picture, is it? And again, the dimension of this problem becomes even clearer when we come to look at this deficit as a share of GDP:

Well, according to Eesti Pank (Estonia's National Bank, see link above):

The average goods and services export growth rate has been fast, although the situation varies by groups of goods and by markets. Although the growth rate of merchandise exports slowed considerably at the end of 2006 and at the beginning of 2007, we are mainly speaking about the so-called transit goods. When we leave aside transit and the subcontracting sector, there are no reasons to assume the competitive ability of other sectors has substantially declined.

This assessment of the situation seems fairly reasonable when we take into account the impact of accelerating producer price inflation and its impact on competitivity. As we can see from the chart below, both exports and imports "peaked" back in May, and this peak in exports occured despite an extraordinarily favourable environment from an external point of view. This is not what should be happening in the case of a "correction". Domestic demand should, of course, be weakening, but exports should be beginning to play an increasing part in maintaining aggregate demand, otherwise this situation simply shows us that Estonia is moving steadily off into a recession, but since the currency cannot adjust, it is not clear where the actual correction to bring her out of the recession again is to come from here. Unless, of course, the recession produces a very sharp deflation in wages, but isn't that exactly what everyone would mean by a hard landing? (ie a strong recession accompanied by a long and sustained period of wage deflation). As I say in my original post I am simply not clear what kind of vocabulary register is being used by many of the participants here. Indeed, I would venture to say that the people who are using this vocabulary are themselves no more clear than I am.

Labour Shortages The Core of the Problem

Again according to Eesti Pank

In spite of slower economic growth, the number of jobs increased, although more sluggishly than before. In the second quarter of 2007, 1.3% more people worked in Estonia year-on-year. Employment increased to 62.9% of Estonia's workforce and the unemployment level fell to 5%.

This is precisely the problem. Even though the growth rate is slowing, the process of labour market tightening continues. This is due to the fact that there is no inbuilt correction mechanism in the labour market due to the shape of Estonia's population pyramid.

What do I mean by this? Well let's look at some more of those charts. Firstly annual unemployment rates in Estonia:

As we can see, these have been steadily coming down since 2000. Now lets look at the numbers of available unemployed:

As we can see the numbers of people registering as unemplyed and seeking work has dropped steadily since the end of 2000. At the same time the number of people working has steadily risen, as is only to be expected given the strong rate of economic growth.

So it is this path which is not sustainable, especially when we take into account that participation rates have also surged strongly.

As we can see the participation rates for older workers is now quite high by international standards, especially when we consider the compartaively low male life expectancy of around 67. Now if we come to look at youth unemployment:

We can again see that the rate of youth unemployment has been declining steadily since the start of 2005. And it is important to remember that these percentages are on a reducing total youth population as can be seen from this, the last chart in this series, which shows how the 0 to 14 age group - or if you like tomorrows young workers and labour market entrants - has been steadily declining as a proportion of the total population since the early 1990s and it is the impact of this steady reduction that is now about to make itself felt.

So there we have it. I basically don't see how - whether you talk in terms of hard landing or soft landing - the Estonian economy is ever going to "correct" unless this structural issue is addressed.

The macro imbalances we are refering to here are substantial, and as I have tried to argue at some length here, they stem from a virtually unique set of circumstances (historically unique I mean, at the present time the underlying dynamic across all the East European EU member states is remarkably similar, with the possible exception of Hungary). Estonia is facing, at one and the same time, a massive inflow of external funds, and a significant reduction in its potential labour supply after years and years of below replacement fertility. Put another way, demand side factors are increasing rapidly, while supply side capacity not only is unable to keep pace, it is actually shrinking (if we think about the number of people of working age).

So there are two problems to correct here, and they are both large and important. Essentially Estonia needs:

a) more labour supply, both skilled and unskilled
b) a lower rate of inflow of structurally distorting funds, whether these be bank credit, remittances, or even (possibly, this needs investigating further) EU funding for projects which Estonia cannot reasonably expect to carry through in the time horizon outlined, given the capacity constraints.
c) more Foreign Direct Investment to create value creating jobs, especially in manufacturing and services areas with export potential
d) increased spending on education and training projects to upgrade the human capital of the existing population

In this post I have been talking about Estonia, but we could just a well have been speaking about Latvia or Lithuania, about Romania and Bulgaria, and - if nothing is done of an international level to address the problem - in the not too distant future about Poland and Ukraine. If we simply sit back and wait for the crunch to come here, then quite frankly it will, and the end product of all that negligence will be much more significant than many seem to currently appreciate.

Wednesday, October 10, 2007

Why Demography Matters, The Romanian Case

Well, the argument on this blog is about to start developing, since events are now (as Claus and I have been trying to suggest since the early summer), moving quite fast. At least in the East European context they are. The day when all the outstanding checks are called due to see if the books balance is getting nearer, and we have all seen only too recently what happens to banks when you need the liquidity to close your accounting for the day and you simply can't find it. Who would ever have thought that this could happen to countries and their stock of people? Well, as we have been saying, the numbers simply don't add up, and in the same way you can't make your bank solvent by "robbing Peter to pay Paul" and frantically shifting money from one account to another, you can't make your continent "population solvent" by massively shifting people around (600,000 from Poland to the UK, 500,000 from Romania to Spain) if there is an underlying deficit (and there is, remember, our current economic growth models are tremendously "people intensive", that is our economies chew up people almost as fast as they burn oil, and with the same consequence, if the supply is constrained the price soars).

At the present point I am busying myself gathering data on Romania, since Romania, for a variety of different reasons, seems to have been singled out by destiny to be the country where all the roads cross, with everything from the global credit crunch emanating from the US sub-prime crisis, to the export dependence of the German economy, to the difficulties presented by tight currency pegs (read Latvia, Lithuania, Estonia and Bulgaria) to the massive migrations of working age population and long term low fertility all coming to a head at one and the same moment, to produce a huge crescendo of cacophony and disorder. And one that may turn out to be contagious, mind you.

People often say that they don't understand why fertility should be so important, or why the hell Edward is so obsessed with this issue. Demography doesn't matter to economics, now does it? The Economist among others have re-assured us on this point time and time again. Aha! Well just key your eyes on those TV screens in the coming months, since all of this is about to be tested, and bigtime I'm afraid. Hopefully I'm wrong.

Incidentally, and following Claus's example, my apologies for the sparse posting of late, but we have been, as you are about to see, somewhat busy.

First off, let's do something unusual for this blog, and start with a currency curve. The euro vs the Romanian leu last Thursday.

Well one thing to remember (since we are measuring the value of one euro in Romanian leu) is that in this peculiar world up actually means down, and hence the point we might like to notice - since the curve goes spiking up - is that the leu took a hell of a hammering on Thursday, that's what it did.

Indeed, since the start of the sub-prime crisis back mid-August the leu has been the worst-performing of the complete batch of 26 key emerging market currencies against the euro. Here's a one month chart as of the end of last week:

Migration the Key

OK, these are currency charts, so what?

Well the point is that in the coming downturn in the global economy (the US is headed slowly into recession, Germany, Italy and Japan already seem to be following suit, and we will see how long the Chinese export model can survive when all the customers are less able to buy) the Romanian economy is going to be particularly exposed, and the reason it is going to be particularly exposed is its fragile demography. Yes, ladies and gentlemen, I kid you not. This IS the situation.

Now the relevant demography here is long term fertility (ie the number of live births since the late 1980s) and the recent huge hemorrhage in out-migration. On fertility we do have reasonably accurate data (more below), but on migration we are scandalously in the dark, despite the fact that one of the key topics we (the "we here" being not the royal one, but "we the economists") need to know about in order to try to get to grips with forecasts and things like that is the level of potential "capacity" in Romania, ie the level of the available labour force. You simply cannot measure potential output if you don't have an accurate measure of how many people you have available. So, to take one 'trivial' example, a central bank cannot carry out a monetary policy analysis, since to facilitate growth, and set and interest rate, you need to know how fast an economy can grow without producing wage inflation and to do this you need an accurate measure of potential labour force growth.

So if we are to make any sort of informed assessment of the future path of the Romanian economy the level of out-migration which has taken place in recent years is a key data point.This I think has to be obvious to anyone with an ounce of economic preparation. And Romania has certainly its full share of out-migration. In this sense the situation is very similar to the Polish one, with the exception that the Polish situation has had a lot more media coverage.

Well.... in order to try and get some data to put online I took the trouble to go and have a look at what Eurostat actually had (the English language version of the Romanian statistical office site, being, unfortunately, a hopeless mess). So here is a chart showing the migration information the Romanian statistical authorities have found themselves able - to date - to provide to Eurostat.

Well, nothing out of the normal here you might think. Romania is simply a country with a very low level of migrant flows. The data is, of course, for net movement, so there could be larger numbers of people coming and going, but still, the order of magnitude difference cannot be that large, or can it?

Now lets take a look at the data provided by the Spanish national statistics office (the INE) for Romanians resident in Spain over the last few years, and compare this data with the official Romanian numbers.

Wow! Well there certainly does seem to be something happening after all. And we should remember that not all the Romanians who have out-migrated since the end of the 1990s have gone to Spain (there are for example a significant number in Italy), although I dare say a fair proportion of them have. The important thing is that we simply don't know. What we do know - and know since the Romanians in Spain (whether they are working legally or not) have an interest (like access to the health system and future amnesties) in registering with the authorities, and indeed the Spanish authorities have (for their own reasons) an interest in maintaining the data in a very up-to-date condition - is that according to the Spanish Padron Municipal electronic-data there were 524,995 Romanians with active and valid id cards for the Spanish health care system as of 1 Jan 2007. These are not just Romanians who are simply passing through, or just might be around somewhere. The municipal registration which lies behind the data is renewable regularly for those without resident permits, and renewing them is how you get the right to have residence later, so this data is VERY accurate.

But we don't know how many Romanians are currently living and working outside Romania in total (or how many Moldovans are living and working in Romania, which is the other side of the issue) and we don't know because the Romanian national statistics office and the Romanian government don't see the issue as important enough to take the trouble to give this the same sort of priority the Spanish government do, despite the fact that Romania is now a member of the EU, and despite the fact that it is considered normal (even by the IMF apparently, reading the staff reports) that Romanian wages should rise rapidly towards European levels. European levels means European levels in every sense of the word.

Of course Romanian wages should steadily rise to the West European level, but the question is how you do it, and the keyword would be "steadily". If you simply increase them by fiat, or because you have no relevant labour supply this won't work. It is not sustainable, and in this way you cannot maintain an internationally competitive economy. You need to raise wages by becoming more productive - per capita - that this the only sustainable way.

Unfortunately the Romanian authorities will live to regret this failing. They will regret it since, as my analysis in this blog post will attempt to demonstrate, in the coming economic crisis in Romania, the outflow of people, and the reverse inflow of money in the form of remittances, when these are coupled with the subsequent distortions in the kinds of economic activity undertaken as a consequence will, unfortunately, prove to have been a very important factor.

(Incidentally, please, will anyone who may be out there with any reasonably accurate data to offer on the migration situation please get in touch.)

Construction, Did Someone Mention Construction?

Coincidentally with all of this Eurostat has just just published its monthly construction production index for all the EU countries. For our present purposes here I reproduce what is perhaps the most significant extract:

Annual comparison

Among those Member States for which data are available for July 2007, construction output rose in seven and fell in seven. The highest increases were recorded in Romania (+26.3%), Slovenia (+17.9%) and Poland (+17.0%). The largest decreases were recorded in Hungary (-14.6%), the United Kingdom (-6.6%) and Belgium (-4.9%).

So construction in Romania is booming. What a surprise! And not only is it booming, it is booming more than anywhere else in the European Union (at least, better put, it was booming while the banks - globally speaking - were still lending money to low-quality creditors for construction purposes, quite how much this activity will now continue as we move forward is exactly the point I am try to make here).

To get a better overall view of the position here's the construction output index for Romania since July 2006:

What we can see is that even though the rate of expansion has slowed noticeably over the last three months, the general expansion in Romanian construction has been significant. Here's the quarterly index for Romanian construction index going back to 2002 which makes the general evolution even clearer:

Now the result of all this frantic construction activity on an economy with tightly constrained labour force capacity (but which is not, please note, inward-capital-flow constrained) was, of course, not long in coming and it was the one which wascompletely to be expected: the price of construction new construction began to strongly surges upwards (readers in the US, does all of this remind you anything, some recent event or other in your country?).

The effect on wages and salaries should also not surprise us too much either, and here I present the index of construction salaries which only serves to confirm our worst suspicions:

Now as we have already noted, and despite the inability of the Romanian government to face up to the implications, there are currently around half a million Romanians working in Spain, and logically these Romanians are not available to work in construction in Romania, so, when you come to think about it, it is hardly surprising that we should find that wage inflation in construction has been enormous. The quarterly year on year chart makes this even plainer. The rate of salary increase in the first quarter of this year, for example, exceeded the astonishing level of a 50% annual rate.

The immigration data is only, of course, on part of the picture here. The other part of the story is the money which the migrants send home, much of which is then diverted into construction. Data on remittance transfers is available from the Romanian national bank, and as can be seen remittance flows have been rising steadily in the last few years, and are currently running at a rate of more than 500 million (or half a billion) euros a month.

This steady upward rise in the volume of inward remittances can also be seen in the following annual chart:

Now one of the central points I want to make here is that this flow in remittances also encourages an increase in the expansion of domestic credit (since the new income can be used to fund the repayments), and this can be seen from in the chart below.

Private credit in Romania is not at preoccupying levels (in terms of % shares of GDP), but it is important to note the steady increase in the quantity of these loans and in the quantity which are foreign currency denominated (normally in euros). In fact foreign denominated loans have been running neck-and-neck with domestic ones in as a share of total credit. This trend of course adds significantly to the vulnerability of the Romanian economy to any currency correction like the one we are beginning to observe in the charts which I presented at the start of this post. And it isn't only the private sector, since foreign currency loan indebtedness is more or less equally distributed between the private and the corporate sector.

General Wage Rises

So OK, wages and costs are going up in construction, but again: so what? Romania is a poor country, wages can rise, they can stand this, can't they? Well unfortunately, they can't, since among other problems the sudden surge in internal demand that all this construction related activity has been producing has created a big hole in Romania's international trade position, and the current account deficit is ballooning, and with this the dependency on international capital glows (and hence on global risk appetite) . Romania is no longer an independent actor here, it depends on the "whims" of its creditors.

But lets look a bit more at some of the general implications of the construction boom. The problem is that the spike in construction wages has not been confined to construction. Here we have the quarterly index for Romanian wages and salaries since 2000. The trend is pretty clear I think. General wage inflation is now an important phenomenon in the Romanian economy.

If we look at the annual rates of increase we will see that from 2001 to the end of 2005 a strict monetary policy from the Romanian central bank was steadily bringing a bad situation under control. But since the start of 2006 this position has once more started to turn round, and has been gradually getting out of control.


So my big argument is that all of this is, in part, driven by a very structurally distorted underlying demography. Let's now take a closer look at this, and to get the ball rolling let's start off with the fertility situation:

Despite the fact that we have a gap in the data during the late 1980s we can see that after falling steadily to replacement level during the 1960s and 1970s, the Romanian fertility rate dropped well below replacement in the 1980s and has been hovering round the 1.2/1.3 "lowest-low" TFR since the start of this century. This is a level in which the population cohort size nearly halves with each generation, so it is no laughing matter.

Now if we look at the drivers of short term fertility, as most of the regular readers of this blog will be aware, they key indicator here is the birth postponement process, and this process revolves around the rising rate of age at first birth of the mothers who have children. In the Romanian case we do not have specific data for this age, but (again thanks to Eurostat) we do have data for mean age on childbirth, and since most Romanian mothers now have only one child this is not a bad indicator.

As we can see, the median age at childbirth has been rising slowly but steadily, and this process is, in part, behind the very low fertility readings which have been registered in Romania in recent years. But at the current level of 27 (and this is an average for ALL births and not just first births, so the first birth median age will be somewhat lower) it is still significantly below the Western European norm, which is around the 30 mark. So more years of very low fertility are to be anticipated as postponement continues, and if Romania hits a major economic crisis in the meantime (which is, I am afraid, a very distinct possibility) then we may get stuck in exactly the kind of low fertility trap which Wolfgang Lutz has so ably identified.

If we now move onto the natural evolution of the Romanian population, I think it is worth starting by taking a look at the live births' chart, because there is, at first sight, a rather strange phenomenon to observe there. This phenomenon is the sudden jump in births in 1969:

Now this jump is undoubtedly the result of one of those rather notorious incidents in Romanian history, the Ceaucescu pro-natality campaign. Perhaps it isn't necessary to say this (it shouldn't be!), but those of us here on this blog who advocate pro-natalist policies along the Swedish or French pattern would obviously wish to completely dissociate ourselves from the type of coercive pro-natalism advocated by Ceausescu and his ilk. What we ARE arguing for is a collective effort on the part of the whole of society (organized collectively and via the state) to transfer resources to those women who would like to have children (Adam Smiths "hidden hand" seems to have gone "missing" at just this point, societies and economies guaranteeing their own reproduction). This is a policy to support choice, but based on the secure knowledge that our collective interest as societies lies in the direction of doing this, and of reproducing ourselves on a stable trajectory. It lies in this direction since otherwise.... well, unfortunately we are just about to find out what the otherwise alternative is in the present Romanian case.

Another import point which comes out of this analysis is the sharp drop in the number of births which took place around the end of the 1980s. To put this in some perspective we might point out that in 1988 some 380,000 children were born, while in 1991 this number was down to 275,000. This is a drop of over 100,000 children a year in just 3 years (and of course the annual rate has subsequently only dropped further). So in 2009 there will be 100,000 young people of 18 years of age LESS than might otherwise have been the case, and the Romanian labour market is BOUND to notice this sharp adjustment, out-migration or no out-migration.

But if we now come to look at the balance of births and deaths, and especially for those which took place in the late 1960s, we will again notice something strange, since we should be able to see that the number of registered deaths rose quite sharply in 1967 and it is plain from the way in which the birth and death curves track one another here that the pro-natalism policy which lies behind the rise in births does seem to have been somehow linked to the dramatic rise in the number of deaths.

Perhaps in comparison with the huge spike in live births the increase in deaths does not seem too significant, but if we look at the chart for deaths a bit closer up, the increase is readily apparent. Deaths in 1967 were up 21,000 on 1966, in 1968 they were up 31,000 and in 1969 they were up by 43,000 over the 1967 figure.

So we have a phenomenon - yet one more time in the Romanian case - which is different from the typical one, in that the numbers of births fell below replacement very early on, and then this fall was followed by a subsequent huge spike which distorted the whole age structure, and then finally the annual number of live births dropped below the number of annual deaths in the early 1990s, and since that time the Romanian population - even ignoring the net loss from migration - has been falling. And if we look at the theoretical population position we will see this confirmed:

Turning to life expectancy at birth, we will see, despite a decline in the early 1990s (which may have been associated with high infant mortality) the headline number is now rising, but it is still low in comparison with Western Europe, and this is going to have important economic implications, not least because it places severe limitations on one of the favourite economic recipes for getting out of the kind of problems Romania is getting into, namely raising participation rates in the over 55 age group. Looking at all this data it may well be that the health of the older Romanian population is just not up to taking this kind of strain. I suspect we are going to hear a lot more of this kind of issue in the East European context in the months and years to come.

Finally, to complete our demographic round up, we could look at the median age of the Romanian population. This age is - at a current level of 36.9 - very young by modern standards, but it is hard to know what interpretation to put on this, since Romania is hardly a "young" population in the sense that Iceland or Ireland are, since the whole structural configuration of the demographic panorama is so uniquely skewed and distorted, and in part the comparatively young median age is a reflection of the comparatively low level of life expectancy.

Now I appreciate that most of you reading this blog are not economists, and hence even if you can see that there is something preoccupying and even deeply disturbing about the demographic profile I have outlined, you can't necessarily see why this is economically important. Maybe you can even see that all this construction activity, and wage push inflation isn't especially positive (and even less so in the wake of the US sub-prime blow-out) but still, why should things be about to go so wrong? Well you need to think about Romania's external position, and the ability of Romania to export its way out of trouble at the same time as the huge wage push inflation means it is experiencing a major loss in international cost competitiveness. Simply put, it is much harder to sell Romanian products abroad. As evidence what I am talking about, here is the chart showing Romania's growing trade deficit:

And here's another one showing the Current Account deficit.

Now this point about the rise of foreign indebtedness brings us directly back to the issue of remittances and construction, and in particular to the high proportion of current mortgages which are being taken out in foreign currency (87% of the total of Romanian mortgages were in foreign currency in July 2007 according to data supplied by the Romanian central bank). This exposure not only to external interest rates, but to the relative values of currencies) makes the individual Romanian household (not to mention their corporate peers) especially vulnerable to any sudden change in the value of the Romanian currency, as the cost of the associated loan repayments would obviously rise significantly. Here's another charts, this time showing the inexorable rise of foreign currency mortgages in Romania.

The point is of course that not only will the repayment on these loans become unbearable for Romanian households if the leu continues its downward decline, the external banks who have these households as clients in their portfolios will also become increasingly exposed (which is why I keep mentioning sub-prime here, it isn't just for the fun of it). Indeed the problem is even worse, since given that these loans are regulated by interest rates established by external banks, the Central Bank in Romania will effectively have little control over monetary policy in the event of a severe downturn. So, simply put, things look bad on all fronts.

And unfortunately - I'm afraid history is just like that, it comes and hits you smack in the face just when you least expect it - just right now we seem to be entering a significant downturn in the global economy whose depth and extent is yet to be determined. And during such downturns the chains normally crack first at their weakest links. Unfortunately for the citizens of Romania their country seems to currently be the weakest link in that 26 key emerging market chain which is chafing at the bit waiting to see who will be the first to fold. So it is to Romania I would be looking if you want to watch that first, initial, blow-out. And the reason why the blow-out will come here, why, "it's the demography, stupid!".