I am reading through Chapter Two of the recent European Economic Advisory Group (EEAG) report on the European Economy 2007. The chapter, which is entitled Macroeconomic adjustment in the Euro Area: the Cases of Ireland and Italy, asks a number of important questions and makes for interesting reading.
The starting point for the chapter is a most pertinent question: six years after the introduction of the common currency, why do important differences in response to the application of a common monetary policy continue to exist across the eurozone? Indeed, on some readings, one might ask why these differences seem to be increasing rather than diminishing, or put another why, why do several significant eurozone economies appear to be diverging, rather than converging. To be sure some differential response among countries to the impact of asymmetric shocks was already anticipated at the outset, but it was always imagined that these differences would diminish and not increase with time. In the same way it was always anticipated that some sort of internal price convergence process would take place, but again this was always expected to settle down with time. Most recent evidence however continues to suggest that there are significant differences between zone economies in the way in which they respond to what is effectively one and the same monetary policy (even though its impact in real interest rate terms obviously differs between zone members as a function of differences in the underlying inflation rate, a fact which, in reality, only poses the same question at one remove, why should we see such a large spread in inflation rates continuing across time?) This conundrum is then the context in which the EEAG sets its comparison.
At the outset of the report the EEAG spell out and make explicit an assumption on which most traditional analyses of adjustment processes within a monetary union have been based.
"In a monetary union among countries with fully flexible prices and wages (and efficient financial markets), an asymmetric demand boom in a country would lead to an increase in the price and wage levels there, reflecting the relative scarcity of its domestic output."
Now this assumption, whilst not being exactly false, does at least seem to be inadequate and in need of revision in the new global economic climate in which we live. One of the key reasons why this "old" approach now seems so questionable (at least in its crude formulation) is the fact that it does not seem to have taken into account the way in which the globalisation of labour and capital supplies may have changed things. The key issue here would be the underlying notion of "capacity" on which it is based, and the extent to which such traditional notions of capacity (and thus of "overheating") remain valid today.
Perhaps the first person to raise this question in any systematic way was Richard Fisher of the Dallas Federal Reserve, and this early speech of his still makes interesting reading. As Fisher says:
Globalization is an ecosystem in which economic potential is no longer defined or contained by political and geographic boundaries. Economic activity knows no bounds in a globalized economy. A globalized world is one where goods, services, financial capital,machinery, money, workers and ideas migrate to wherever they are most valued and can work together most efficiently,flexibly and securely.
Now, as Fisher asks, "Where exactly does monetary policy come into play in this world?" Well let's see:
The language of Fedspeak is full of sacrosanct terms such as “output gap” and “capacity constraints” and “the natural rate of unemployment,” known by its successor acronym, “NAIRU,” the non-accelerating inflation rate of unemployment. Central bankers want GDP to run at no more than its theoretical limit, for exceeding that limit for long might stoke the fires of inflation. They do not wish to strain the economy’s capacity to produce. One key capacity factor is the labor pool. There is a shibboleth known as the Phillips curve, which posits that beyond a certain point too much employment ignites demand for greater pay, with eventual inflationary consequences for the entire economy.
I cite Fisher at length here, since he does seem to be directly challenging the kind of assumption - as spelt out by the EEAG - which I am drawing attention to. As he goes on to say: "Until only recently, the econometric calculations of the various capacity constraints and gaps of the U.S. economy were based on assumptions of a world that exists no more". A world that exists no more. Please note. (A good summary of the kind of argumentation that lies behind Fisher's approach can be found in the central chapter of the Dallas Fed 2006 Annual Report - Globalizing the Knowledge Economy).
Now, to qualify what I am saying a little, the issue is not that individual domestic economies are not subject to price and wage rises following surges in domestic demand, but that they are not subject to these to anything like the degree they used to be, and unless allowance for the extent of this change is made in the econometric models used, then traditional notions of "capacity" are likely to lead you well wide of the mark in looking at the impacts of monetary policy changes, since capacity itself has become much more elusive and elastic, and it is this very elasticity - ie the capacity for local economies to draw on large pools of underutilized labour, and over long distances, and avail themselves of the increased (and normally cheaper) supplies of capital which are available through global financial markets (and of course in the eurozone context the European capital markets themselves) - which means they are able to respond to rapid increases in demand without the normal wage and prices pressures coming into play to anything like the extent that they once did.
In this post (which will be in three parts) I will follow the EEAG lead, and look at the two countries individually (Ireland and Italy), treating them to some extent as case studies to see what may be learnt (and a reflection on this will constitute the third part of the post). Firstly Ireland.
Europe's Celtic Tiger?
According to the report:
"Ireland entered the euro area well into a sustained period of economic expansion marked by profound changes in the structure of the economy and its place in the global economy".
As can be seen from the chart below, Ireland has enjoyed quite high rates of GDP growth in recent years:
This process has seen a rapid rise in per-capita incomes, and a dramatic fall in the level of unemployment. Interestingly this rapid economic development upswing coincided with a process known to economists as the demographic dividend, as I note in this post on "the Celtic tiger".
The process of monetary union produced a strong monetary stimulus in the Irish context, as the report indicates:
The most apparent and controversial source of macroeconomic imbalance for Ireland has instead been the strong monetary stimulus since the end of the 1990s, when European monetary policies became strictly coordinated in the last stage of nominal convergence before the introduction of the euro. (Soon afterwards, the monetary stimulus was compounded by a weakening currency.)
Now as the report goes on to say, this monetary stimulus was required to meet the needs (low growth rates, sluggish internal demand) of other eurozone members (among them Italy, but also importantly Germany), even though it was arguably (at least in classic terms) inappropriate in the Irish context:
A relatively loose monetary stance was motivated by the cyclical conditions in the euro area as a whole, but arguably inappropriate forIreland: It created undue demand pressures in the Irish economy.
Now on the traditional account, the presence of such "undue" demand should have lead the Irish economy to "overheat", producing in its wake a growing inflationary wage price spiral., and indeed on conventional measures this was what was happening. But does this view correspond to the Irish reality?
True the Irish economy has had noticeably higher inflation rates than the EU average in recent years, but this process has hardly spiraled out of control.
Indeed if we look at the chart below, we will see that hourly labour costs in manufacturing in Ireland are still in the lower end of the eurozone range:
So - despite the fact that there has been a substantial rise in wage costs (from a very low level) if we allow for some measure of eurozone price and wage convergence over time, the general impact does not seem to have have been to force up prices and wages to such an extent that the Irish economy became uncompetitive. Curiously, if we look at the official Irish CPI, we will find that this actually trended down during the first years of the century, moving from an annual rate of 4.7% in 2002, to 3.5% in 2003, 2.2% in 2004, 2.4% in 2005, and only turning up again to 3.9% in 2006, a move which seems to have been associated with the fact that mortgage interest payments are included in the Irish CPI calculations. This hardly constitutes strong evidence of an inflation-push process fuelled by overheating, and especially not in view of the very rapid rates of GDP growth which Ireland was experiencing. So what happened?
Well, in a nutshell, Ireland got immigration, bigtime, and the combination of this with a ready supply of cheap capital seems to have maintained Ireland on what might be considered to be a sustainable path. The general trend in migratory flows can be seen in the chart below, but as an indication it could be noted that in 2006, 86,900 people immigrated into Ireland, the equivalent of 2.1 percent of the Irish population and 4.1 percent of the labour force.
As the February 2007 reprt of the Economist Intelligence Unit notes:
Irelands labour force totalled 2,178,100 in the third quarter, a year-on-year increase of 4.4% or 92,000. This is extremely high by international standards the most recent EU-wide data relate to the second quarter of 2006 and put Irelands rate of labour-force expansion at 4.6%, almost four times the EU average of 1.2%. Immigration has been a crucial factor in this. Demographic change accounted for three-quarters (or 69,100) of the Irish labour forces growth in the year to the third quarter of 2006, and immigration accounted for 70% of this figure. The remaining, non-demographic increase in the labour force reflects continued increases in the participation rate, which rose from 63.2 to 64.1 between the third quarters of 2005 and 2006.
So in part recent growth in the Irish economy has been facilitated by growth in the domestic availability of labour (whether through the arrival in the labour market of Ireland's still relatively numerous young cohorts - given Ireland's comparatively strong recent fertility levels - and through an increase in participation rates, both of these factors forming part of the underlying demographic dividend process) and via the arrival of migrant labour on a very large (indeed almost unprecedented) scale.
Returning to the EEAG report, they go on to argue:
In principle, a strong demand expansion should have created a severe labour shortage.But the booming economy stimulated a strong migratory inflow, with two major effects: first, the increasing supply of labour contained upward pressures on wages somewhat, especially in low-skill occupations.8 Second, the additional workers in the economy raised aggregate demand, reinforcing the expansionary macroeconomic stance for the economy as a whole. Since the availability of jobs acts as a strong driving force for migratory decisions, a sustained economic boom created incentives for further migration.
So the low interest rate environment created a migratory flow which had an important multiplier effect (in classic Keynesian terms). And indeed the report concedes that while:
"adjustment (to the monetary shock) seems to have worked as predicted by theory.... the....overall expansionary policy mix caused real appreciation, although adjustment through wages and labour costs was arguably contained because the strong migratory inflow reduced excess demand in the labour market.
My feeling is that the whole classic account of monetary shock adjustment is rather more challenged by all of this that the EEAG seem to realise, but still.
On the other hand the capital required to enable domestic output to expand so rapidly was readily available internationally, and at remarkably low rates of interest given the presence of the single size for all euro environment. Some evidence of the extent to which the Irish business sector - and indirectly Irish mortgage borrowers - have had access to global found in the graph below, where it can be seen that there has been a very rapid growth in foreign borrowing by Irish banks to finance domestic lending. Since the end of the last century bank borrowing from abroad to on-lend to Irish residents has soared from 10% to 41% of GDP.
Ok, so what I think is now clear is that if Ireland has had an unprecedented economic boom in recent years, it is in part because the "spare" capacity was available (elsewhere). But what about the growth itself. What were the structural drivers?
Well first and foremost, domestic consumption (and the associated consumer indebtedness) has been a significant component.
Private consumption has been growing strongly in recent years, but beginning in the third quarter of 2006 it has been easing back (showing a year-on-year increase of 4.5%, its lowest level since the final quarter of 2004), a sharp drop from the 7.1% recorded in the second quarter.
Well clearly the boom in the housing sector has been an important part of this process. As the Economist Intelligence Unit note (2007) "Residential property prices in Ireland have risen more rapidly than in any other developed world economy over the past decade". Using conventional terminology they then go on to say:
"However, given that the increase in supply of new housing has been just as phenomenal, such rises appear unjustified by the fundamentals (the number of annual housing completions is almost five times that of the early 1990s. this compares with largely static output in the euro area and the UK)."
Really I feel this type of response simply begs the question, whether or not this housing surge is justified by "fundamentals" is in part a question of determining just what these fundamentals really are, and about how well we are actually measuring them, not to mention how fluid they may have become, and whether they are not now something of a moving goalpost, at least in the Irish context.
Certainly, as the EIU suggests, housing growth has been "phenomenal". One indication of this is the the fact that almost a third of the current housing stock has been built since 1990, so Ireland now has the lowest average age of dwellings in the EU (something which we ought not to find *so* surprising since with a median age of only 34 Ireland is still far and away the youngest society in the EU). Annual completions have been running 3.5 times what they were a decade ago and the country has the highest per capita building rate in the EU. But the real issue is, going forward, just how sustainable is all of this? But first,let's take a look at a chart (below) which identifies the impact on house prices of all this construction activity:
Now one of the most interesting things to note from the chart is that the largest spurt in price increases seems to have taken place in the 1997-99 period (ie before the official introduction of the euro, and before perhaps the more flexible supply of migrant labour and capital became available). A second (smaller) spike seems to have occurred after 2003, but it has not been anything like so dramatic. On the back of the most recent round of interest rate tightening from the ECB housing activity in Ireland has been slowing notably and indeed the average price paid for a house in Ireland in March 2007 was 2,007 euros less than the average price paid in February according to the latest edition of the permanent tsb /ESRI House Price Index. This is equivalent to a decline in national prices of 0.6% month on month. This is the first reduction in national house prices since January 2002, when prices declined by 0.9%. In the first quarter of 2007 (January to March) prices nationally decreased by 0.5% as compared with a growth of 3.5% in the same period of 2006. However on a year-on-year basis the average price pay for a house in Ireland was still 7.4% higher in March 2007 than the average price paid in March 2006. The big question is of course what happens next.
This is not the place to enter a debate as to whether what has been happening in Ireland constitutes a bubble (I have some observations on the Indian property situation which are not without relevance here), or indeed of the extent to which such price growth as has occurred represents a distortion from "fundamentals", since this in large part depends on the extent to which fundamentals have changed, and this is precisely what we are still trying to determine. Evidently, with Irish consumers now the most indebted in the OECD, any sustained decline in property prices would make its presence felt via the well-known "wealth effect".
But just how likely is such a significant "correction". In large part the "sustainability" of recent housing activity depends on the future course of interest rates at the ECB, and this is again beyond the scope of the present post, but suffice it to say that the current raising cycle may well peak (due to its impact on the larger economies like Germany and Italy with different structural characteristics to the Irish one) much sooner than many imagine - in the last quarter of 2007 or early 2008, and we may well then see another round of "easing", if so the associated fall in interest rates may well serve to place a platform under house prices in Ireland, and indeed the show (in perhaps a more modest form than hitherto) may well go on. Certainly there are no good reasons at this point to exclude this possibility.
In addition it should be noted that Irish government finances have enjoyed a strong positive balance in recent years, and that any slowdown in the housing sector can - to some extent - be counteracted by the application of a countercyclical deficit. One indication that such an approach - in the eventuality that it may prove necessary - may not be far from government thinking may be found in the most recent National Development Plan. This is projected to cost 183.7bn euros over a seven-year period (2007-13). The NDP is aimed at tackling what are widely acknowledged to be serious infrastructure deficits. In addition - and for the first time in an Irish NDP - the package of measures includes current spending (86bn euros). This current spending, which will be focused chiefly on social expenditure, is dependent on tax revenue remaining buoyant (the governments plan is premised on average growth rates of 4%-4.5% over the period). As a percentage of GNP, capital spending will increase from 4.7% in 2006 to an average of 5.4% over the term of the seven-year plan.
Finally, it would not be appropriate to leave this review without at least some comment on Ireland's external balance position. In the first place it should be noted that Ireland's current account is un-typical, in the sense that, while there is a small trade surplus, the current account balance is seriously in the red. This is largely due to the strong presence which FDI has had in Ireland over the last decade or so, and the strong consequent outward funds flow associated with the repatriation of profits (which has the further un-typical consequence that GDP is systematically lower than GNP.
Evidently Irelands current-account deficit continues to widen. In the first nine months of 2006 a cumulative deficit of 5bn euros was recorded, a figure which was up sharply from the 3.9bn euros recorded a in the same period one year earlier. The third-quarter deficit totalled 1.2bn euros, an increase in 46% over the deficit in the same period of 2005. The increasing current-account deficit is largely a reflection of changes in the income balance, since the merchandise surplus was almost unchanged year on year at 7.6bn euros, while the services deficit actually decreased. However, this narrowing in the services deficit was more than offset by a widening of the income deficit from 5.4bn euros in the third quarter of 2005 to 6.8bn euros in the same period of 2006.
This situation in part reflects an increase of 15.9% in profit outflows generated by foreign-owned business based in Ireland and of a 41.4% increase in portfolio investment income outflows. As with other deficit countries (Spain, the UK, the US) there is no doubting the existence of such "imbalances", the real issue is their significance. As Claus and I have been repeatedly pointing out (and here), if median ages to some extent govern the dynamics of domestic consumption, and some high median age societies (Germany and Japan, for example) are condemned to running current account surpluses if they wish to maintain growth, then logically others are condemned to run deficits.
So, and summing up, what we seem to have in Ireland is a very interesting test case for the validity of the old versus the new models of capacity. On the face of it I would argue that global capital and labour flows have had a significant - and unexpected - impact on the path of the Irish economy, a development from which there is much to be learned. Let's just hope that over at the ECB they are listening.
NB. This post is the first installment of a much longer review post. More to come. Next stop Italy.