Figure1b shows these same financial deficits/surpluses, but only for the govern- ment sector and the private sector12and distinguishing between the external sur- plus and deficit groups. Indeed, for analytical purposes, the accounts are shown for two groupings of euro area countries,13grouping together countries that had run
-600 -400 -200 0 200 400 600
-600 -400 -200 0 200 400 600
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
households non-financial corporations financial corporations government euro area
-600 -400 -200 0 200 400 600
600 400 200 0 200 400 600
government of external deficit group government of external surplus group private sector of external deficit group private sector of external surplus group euro area
Surplus group Deficit group
(a) (b)
Fig. 1 Euro area net lending/net borrowing(four-quarter sums; EUR billions).aBy sector.bBy country grouping for the public and private sectors.SourceEurostat and ECB. Note The net lending/net borrowing shown in the charts of this box has been adjusted, for convenience, so as to exclude ‘‘acquisitions less disposals of non-financial non-produced assets’’ (in order to avoid the distortions caused by the large proceeds from the sale of UMTS mobile phone licences in 2000)
11 The net lending excludes however, holding gains/losses or other write-offs on assets.
12 Defined here as the sum of all non-government sectors (thus including public corporations).
13 This type of presentation was first used in ‘‘The financial crisis in the light of euro area accounts’’ (ECB2011a). The grouping aggregates are obtained by simple aggregation of national data, while maintaining additivity to euro area totals, by way of allocating any difference relative to the euro area totals (stemming mostly from intra-euro area balance of payments asymmetries) to each grouping on a pro rata basis. No further consolidation is conducted (which is broadly appropriate as the EAA are mainly compiled on a non-consolidated basis).
current account surpluses (external surpluses) over a period of five years ending with the onset of the financial crisis in 2007 (‘‘external surplus group’’—Belgium, Germany, Luxembourg, the Netherlands, Austria and Finland) and separately those that ran current account deficits (‘‘external deficit group’’—Ireland, Estonia, Greece, Spain, France, Italy, Cyprus, Malta, Portugal, Slovakia and Slovenia).14 The criterion used here to assign countries to each group is chosen for illustrative purposes, to work out some common stylised facts that can be observed in the boom period. Each of the groupings are rather heterogeneous, for instance, com- prising countries with very large external deficits or surpluses, while others have current account positions that are close to balance. Countries also differ consid- erably concerning other indicators (such as fiscal position, presence of boom-bust housing market cycles, etc.). In addition, the composition of the group is obviously closely tied to the reference period and would change over time. Germany, for instance, would have been in the ‘‘external deficit group’’ in case a similar exercise had been conducted at the beginning of the century, while Italy and France would have been in the ‘‘surplus group’’ at that time: this, in itself, underscores the important point that corrections and reversals of imbalances within Monetary Union do occur over time.
Taking such a grouping-of-countries view, Fig.1b highlights the pronounced increase in financial deficits of the private sector in the external deficit group during the boom years, matched by stable and ample private sector surpluses (as well as by sharp reductions in government deficits) in the external surplus group.
From 2008, the financial crisis triggered an abrupt reduction of the financial deficits of the private sector in the external deficit group, which turned into sur- pluses mid-2009. At the same time, in the external surplus group, the private sector surpluses increased further. In the absence of any significant improvement in external balances (the line ‘‘euro area’’ in Fig.1), these mounting private sector surpluses had their counterpart in generally higher government deficits.15
Furthermore, as the external deficit group, taken as a whole,16did not improve their fiscal situation sufficiently during the boom years (given that they still had an
14 Greece, Cyprus, Malta, Slovakia, Slovenia and Estonia data are included over the whole period studied, despite their having joined the euro area only progressively (‘fixed composition’
presentation).
15 It should be noted that this fundamental accounting constraint does not, in itself, indicate the direction of causality, i.e. whether the government deficits resulted from increased private surpluses/saving or, alternatively, whether the latter reacted to increased government deficits.
16 Some countries in the external deficit group (such as Spain or Ireland), however, recorded government surpluses at the height of the boom. Their current account deficits therefore reflected private sector dissaving and lagging competitiveness.
In Need of Sectoral and Regional Rebalancing in the Euro Area 75
overall deficit of 1.4 % of GDP in 2007), their public finances became seriously impaired by 2009–2010, and in need of substantial and immediate corrective measures. This contrasts with governments in the external surplus group that used the boom period to turn their overall deficit into a surplus (in 2007), although this did not prevent the later occurrence of deficits that were also in need of correction.
In total, the gradual but ultimately substantial increase of the gap in external balances between the two groupings that emerged prior to the recession of 2008 failed to reduce noticeably thereafter, during the recession and the following recovery. It hence failed to respond to the considerable adjustment in the private sector balances that seemed largely compensated, or neutralised, by matching movements in government deficits.
A more complete sectoral decomposition of the differences in private sector balances between the two country groupings can be observed in Fig.2. During the crisis, starting from the far lower levels reached at the height of the boom, the net lending of households increased more in the external deficit group than in the external surplus group. Financial corporations’ surpluses (mostly their retained earnings) were significant in both country groupings, but increased slightly more in the external surplus group in the wake of the crisis, after having declined there at the peak of the boom.
Overall, the heterogeneity between country groupings appears most pronounced in the case of NFCs. First, whereas the NFCs in the external deficit group main- tained a traditional17 net borrowing position throughout the period, those in the external surplus group experienced atypical long-lasting net lending positions as from 2003, positions of the kind that can be observed during recessions or that can be associated with strong foreign direct investment abroad. Second, the expan- sionary financial balances of NFCs in the external deficit group turned around earlier (compared to external surplus group) at the start of the crisis, with their net borrowing position peaking in the third quarter of 2008. In contrast, in the external surplus group, the peak was only reached six months later in the first quarter of 2009: it is the crisis itself that pushed corporates in this group from a surplus to a deficit position, essentially via a steep reduction in their retained earnings.
17 It is customary to assume that firms borrow or raise equity so to fund investment (‘K’) from households (directly or indirectly) that save for future consumption. But an alternative growth model is when firms generate sufficient retained earnings to fund themselves (in aggregate) all the required investment. In this case, household wealth (and therefore their capacity to fund future consumption) still increases: not via their own net saving, but via holding gains on equity held (stemming from the net saving of corporates). Note that these two growth models would be reflected in the national accounts identically if the accounting rules were to assign the retained earnings of corporates as investors/households’ income, which is currently not the case.
4 Two Views on the Increasing Regional Imbalances
The growing imbalances between the two country groupings in the run-up to the boom can receive two very different interpretations. According to one view, the imbalances reflect increased financial integration and the easier cross-border cir- culation of savings within the Monetary Union. In this somewhat benign view, imbalances would be perceived as allowing an optimal allocation of savings across more profitable investments prospects, assuming a sufficiently efficient interme- diation process carried out by financial institutions and markets alike.
Feldstein and Horioka had found (1980) that national investments and savings tended to be highly correlated across countries, and interpreted this as evidence that world capital markets were not well integrated, somewhat in contrast to a first impression (i.e. the ability to shift funds around easily): the so-called ‘‘Feldstein- Horioka puzzle’’. To explain this, they hypothesised that portfolio preferences and institutional rigidities impede long-term capital flows—short term capital mobility was not concerned, the authors argued, as revealed by the fact that short-term covered interest rate differentials are negligible. This optimist view of the growing external imbalances posited that national saving and investment tended to be more and more disconnected within the euro area, thus bringing an end to the Feldstein- Horioka puzzle. Blanchard and Giavazzi (2002) for instance suggested such a process to be at work in the case of Portugal and Greece.
Another ‘‘pessimistic’’ (or ‘‘realistic’’) view was that the growing external imbalances reflected the impact of local demand booms and supply rigidities, as well as associated distortions in competitiveness. In this view, the circulation of savings from surplus to deficit countries is not possible without pushing prices and wages away from initial equilibrium, unless goods and labour markets are fully
-10 -8 -6 -4 -2 0 2 4 6 8 10
-10 -8 -6 -4 -2 0 2 4 6 8 10
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 government
financial corporations non-financial corporations households total economy
-10 -8 -6 -4 -2 0 2 4 6 8 10
-10 -8 -6 -4 -2 0 2 4 6 8 10
government financial corporations non-financial corporations households total economy
(a) (b)
Fig. 2 Net lending/net borrowing by country grouping. (four-quarter sums; percentages of GDP).aExternal surplus group.bExternal deficit group.SourceEurostat and ECB.NoteThe net lending/net borrowing shown in the charts of this box has been adjusted, for convenience, so as to exclude ‘‘acquisitions less disposals of non-financial non-produced assets’’ (in order to avoid the distortions caused by the large proceeds from the sale of UMTS mobile phone licences in 2000) In Need of Sectoral and Regional Rebalancing in the Euro Area 77
integrated (without frictions). Hence, the free circulation of savings between countries requires in practice commensurate relative movements in prices/wages (or a change in demand stemming from changes in preferences).
This second view emphasises that the correlation between savings and invest- ment tests as much the integration of the goods markets as it tests the integration of the capital market. This is, in fact, a joint test. A convenient way to see this is to recall that the savings/investment equilibrium is simply the reverse of the goods market equilibrium: it is not a ‘‘market’’ on its own, where for instance interest rates would be fixed (they are fixed on the money market). For saving to circulate between countries, i.e. to be ‘‘exported’’ from country A to country B, there is a need for net exports to be positive in A and negative in B. In turn, this requires an adjustment of prices/wages so that global demand starts favouring products of country A. In the absence of positive net exports, any financial investment into B by willing entities of A (say an equity purchase, a bank loan) is automatically matched by a financial investment from B to A (say an interbank deposit, or if no entities in B is willing to invest in A, a so called ‘Target II’ balance18). Conversely, if B has a deficit against A but investors in A do not wish to continue purchasing claims on B, what happens? In a common currency zone, the deficit of B will continue as long as demand is not compressed (e.g. fiscal contraction, banking constrains), or prices/wages are not adjusted. Financing is simply ensured via the Euro system (‘Target II’ balances). In a two currency regime, there will be devaluation of currency B vis-à-vis A that will bring about three adjustments altogether: (a) reducing prices/wages, (b) reducing demand and (c) stimulating asset purchases by entities in A (attracted by the discount).
In order to better understand the origin of sectoral imbalances in the euro area, we examine the other elements of the accounts, in particular the regional differ- ences in savings and investments.
5 Regional Patterns of Savings and Investments
Useful insight can be gained from the analysis of surpluses/deficits by looking at the dynamics of the two main components of net lending/net borrowing, namely investment (gross capital formation) and saving (including net capital transfers).
Fig.3a shows the dynamics of the differentials between groups in both the saving ratios (i.e. the ratio of domestic saving to GDP in the external surplus group minus that in the external deficit group) and the investment ratios. These explain the dynamics of the gap in external balances between the external surplus group and the external deficit group.
As can be seen from the chart, the gradual but ultimately substantial increase of this gap in external balances prior to the recession of 2008 was driven mainly by
18 See Box 4 ‘‘Target2 balances of national central banks in the euro area’’ in (ECB2011b).
increasing domestic saving differentials and to a lesser extent by increasing investment differentials (through ever higher investment ratios in the external deficit group). This observation is reinforced over the full observation period of 2002–2012.
Figure3b focuses on the saving rates (national savings).19 It shows the rapid expansion of the saving differentials until 2007 resulting from slightly falling saving ratios in the external deficit group standing in stark contrast to the pro- nounced increase in the external surplus group. In addition, the chart shows the sectoral contributions to the change in saving ratio differentials. The divergence in saving behaviour between the two country groupings clearly largely originated in the NFC sector, where the saving differential rose until 2008. This reflects the fact that the ratios of NFC saving to GDP in the external surplus group increased persistently throughout the five years to 2008, while at the same time they edged down steadily in the external deficit group. By contrast, the differential in household saving remained more stable over time: it fell moderately from 1.5 % of GDP in 2000 to close to zero in mid-2002, and increased again up till 2005, remaining at this level until 2008. As well, the differential in government’s and financial corporations’ savings hardly changed.
During the recession of 2008–2009, the saving differentials decreased to some extent for NFCs, as corporate saving contracted more in the external surplus group than in the external deficit group, and to a lesser extent for households with saving increasing more in the external deficit group than in the external surplus group.
contribution of the saving ratio differential (external surplus group minus external deficit group) contribution of the investment ratio differential
net lending/net borrowing - external surplus group net lending/net borrowing - external deficit group
14 16 18 20 22 24 26 28
14 16 18 20 22 24 26 28
NFC saving differential (contributions) households saving differential (contributions) financial corporations saving differential (contributions) government saving differential (contributions) total saving - external surplus group total saving - external deficit group
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
(a) (b)
4 6 8
-6 -4 -2 0 2
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Fig. 3 aDifferentials between external surplus group and external deficit group in saving and investment ratios. (four-quarter sums; percentages of GDP).SourceEurostat and ECB.NoteThe saving ratio differential includes net capital transfers.bDifferentials between external surplus group and external deficit group in sectoral saving ratios. (four-quarter sums; percentages of GDP).SourceEurostat and ECB.NoteNegative sectoral differentials (i.e. high saving ratios in the external deficit group) lead to entries below the line for ‘‘total saving—external deficit group’’
19 Savings of all sectors (as a percentage of GDP), rather that the savings rate of households (saving to their disposable income).
In Need of Sectoral and Regional Rebalancing in the Euro Area 79
These significant symmetric movements in saving during the recession were subsequently partially reversed. Overall, the corporate saving differential remained fairly large by mid-2012 in between the two groupings.
Whereas there had been only few divergences in government savings between the two country groupings before 2007, these became notable thereafter: during the recession, government saving fell faster and more steeply in the external deficit group. This drift was not corrected, but compounded by the stronger rebound since mid-2010 in government savings in the external surplus group, and gross saving turned again positive there in the 12 months to the second quarter of 2011.
As a consequence, by mid-2012, the differential in government saving con- tributed to the national differential between saving as much as, or somewhat more than, that in corporate saving.
6 Differentials in Corporate Margins and Profits
A main driver for the decline in retained earnings and the associated high deficit position of NFCs in the external deficit group is their lower profitability, as measured by their ‘margins’: gross operating surplus to value added (see Fig.4).
These margins were at similar levels of around 38 % in the two country groupings until 2004, but started to diverge thereafter, increasing to a maximum of 43.7 % at the end of 2007 in the external surplus group, while they fell in the external deficit group. This opened up a gap of almost 6 % points, which narrowed temporarily during the 2008–2009 recession, but started to widen again during the subsequent recovery. As of the second quarter of 2012, NFC margins generally remain
34 36 38 40 42 44 46
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 external deficit group
external surplus group Euro area
Fig. 4 Ratio of the gross operating surplus to value added of NFCs. (four-quarter averages; percentages).
SourceEurostat and ECB
depressed in the external deficit group, standing 3.5 % points lower than in the external surplus group.
The main reason for the lower corporate margins of the external deficit group is to be found in the far larger increase in wages paid by businesses in the period from 2000 to 2011 (see Fig.5), an increase over and beyond what would have been justified by stronger output growth (higher productivity and employment gains) in those countries. Indeed, any change in total compensation of employees can be exactly decomposed into output growth in volume terms and changes in unit labour costs. In the external deficit group, the latter rose by 28 % in the ten years to 2010, compared with an increase of less than 11 % in the external surplus group.20
This gap thus reflects wage growth in the external deficit group over the past ten years that was excessive in comparison with that in the external surplus group, leading to a loss of competitiveness. This compressed corporate margins in external deficit group, as businesses could not pass on all cost increases in full, especially in the case of exposed tradable goods and services.21
Looking ahead, assuming that the compensation of employee growth observed over 1999–2011 would be consistent in future with inflation below but close to 2 % over the long run, and thus does continue on this trend, the decisive issue is
1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2
0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 exterrnal deficit group
external surplus group All euro area (RHS)
Fig. 5 Compensation of employees paid by NFCs.
(four-quarter sums; EUR trillions).SourceEurostat and ECB
20 Weighted by GDP.
21 Even if the higher nominal wage increases in the external deficit group reflected, merely or mostly, higher domestic inflation, this nonetheless caused a deterioration in competitiveness, and thus additional pressures on the margins of businesses exposed to international competitors (including those in the other grouping of the euro area).
In Need of Sectoral and Regional Rebalancing in the Euro Area 81