Real interest rates in the EMU: Facts and implications

by Michael G. Arghyrou1

Economics Section, Cardiff Business School

The European Central Bank (ECB) currently faces probably the most challenging financial and economic environment since its inauguration. Global inflation pressures compromise its institutional price-stability objective while the risk of the credit-crunch crisis spilling over to the euro-zone’s economy is real. The challenge for the ECB is to determine an interest rate policy maintaining low inflation without causing a euro-zone recession. In this respect, the ECB is not different to other major central banks, such as the Federal Reserve Bank or the Bank of England. What distinguishes it from the latter is that it pursues its objectives on behalf of different sovereign states with heterogeneous real rigidities, business-cycle stages and exposure to the credit-crunch risks. These are significantly complicating factors, as they may result in asymmetric transmission of the single monetary policy and hinder the ECB’s ability to meet objectives defined in terms of EMU aggregates. If persistent, these may then invite political pressure on the ECB and undermine confidence in its ability to deliver effective monetary management. Let us give an example to make the argument clear.

Let us assume that the EMU is divided into two groups, a low-income, fast-growing, rigid Periphery and a high-income, slow-growing, flexible Centre. To control inflation in the rigid Periphery, the ECB would implement interest rate increases at the cost of unnecessary output volatility in the Centre. Alternatively, it can opt for an interest rate policy ensuring output smoothing in the Centre at the risk of macro-imbalances in the Periphery (e.g. increasing current account deficit and high asset-price increases). Let us assume that the ECB opts for the second alternative due to the Centre having a larger economy than the Periphery, in which case its policy keeps EMU aggregates near their targeted values. Macroeconomic adjustment in the Periphery is then left to the flexibility-promoting effects of the Periphery’s structural policies; and the demand-mitigating influence of Balassa-Samuelson effects caused by higher relative inflation and real exchange rate appreciation. As long as the global economic outlook is benign and liquidity provision is abundant, inter-temporal constraints in the Periphery are not too binding and macro-adjustment can take place smoothly. Overall, in good times this policy keeps all three parties (ECB, Centre and Periphery) largely content. It also summarises, broadly speaking, developments in the euro-zone over the past decade.

This policy, however, carries a risk. This is that the Periphery will not promote structural policies actively enough; and/or its high growth rates will cause over-optimistic risk assessments, unsustainable increases in asset prices, investors’ and consumers’ exuberance and excessive internal/external borrowing. Imbalances in the Periphery may then become so pronounced in size to create bubble-like conditions, rendering its economy vulnerable to the risk of a boom-and-bust. The trigger may be a liquidity shock, such as the current credit crunch, suddenly increasing the cost of borrowing; and/or a global inflation shock, similar to the present one, prompting the ECB to increase its interest rates abruptly. If the latter does not do so, it will be compromising its price-stability objective, thus risking dissatisfaction in the Centre. If it does it may cause a strong economic downturn and dissatisfaction in the Periphery. If, finally, the ECB takes a middle-way course, it risks missing its EMU-average objectives, causing dissatisfaction in, and inviting political pressure from, both groups. Overall, in tough times heterogeneous real rigidities, differential growth rates and pronounced imbalances may cause problems to all three parties. If persistent these, as argued above, may undermine confidence in the effectiveness (or even the legitimacy) of the single monetary policy.

How close are we to the unfavourable scenario described above? The ECB has recently signalled its intention to tighten monetary policy by increasing, for the first time in a year, its key main refinancing operations interest rate by 0.25% to 4.25%. This is a clear signal that the ECB regards the risk of inflation to be a more serious threat than the risk of recession for the euro-zone’s economy. Two questions now arise. First, are higher ECB interest rates likely to lower EMU-average inflation given the present global inflation pressures? Second, are they likely to trigger a recession among the EMU’s high-growth members? A variable whose movements convey useful information relating to both questions is the real interest rate differentials (RIRDs) of individual EMU members against the union’s average. This is so because the real interest rate channel is the main channel through which monetary policy is transmitted to the euro-zone’s individual economies (see Clements et al, 2001). As such, a necessary (though not sufficient) condition for uniform transmission of the single monetary policy is that national RIRDs against the EMU average are mean-reverting and display similar persistence patterns. If the opposite is true, shifts in the ECB interest rate would pose EMU members with asymmetric monetary shocks causing differential and potentially costly inflation, output and asset prices’ responses, compromising the EMU-average price stability objective (see ECB, 2003). Furthermore, if real interest rates within the EMU are bound to converge, large RIRDs are indicative of unsustainable macroeconomic imbalances; in which case their size at any moment in time is a good proxy for the risk of recession.

In a paper recently published by the Journal of International Financial Markets, Institutions & Money, Arghyrou, Gregoriou and Kontonikas examine real interest rate convergence against the EMU average in the EU25 area for the period 1996-2005. The authors find evidence of convergence for the majority of countries, including most new EU members. Convergence, however, is found to be a gradual and non-uniform (in terms of short-run dynamics) process, subject to structural breaks falling close to important economic events most prominently (though not exclusively) the euro’s launch in 1999. Furthermore, among EMU members, convergence is rejected for Greece and Spain, for which increasingly negative RIRD-trends are observed in recent years. Interestingly, Ireland, Italy and Portugal, are found to have diverged from the EMU average during the early euro-years but have converged to the latter since 2004. Finally, among non-EMU members, convergence is rejected for Latvia, for which a negative non-converging RIRD is found, and Hungary, Poland and the United Kingdom, which present positive, non-converging RIRDs.

Arghyrou, Gregoriou and Kontonikas analyse individual RIRD movements in terms of time-varying risk-premiums and productivity gains. They conclude that for the majority of EMU countries, where convergence is upheld, the steady-state costs of losing monetary independence should in principle be not too high; and that the majority of the new EU countries are significantly closer to joining the euro than ten years ago. However, these optimistic implications come with one important caveat, namely that non-uniform speeds of convergence imply that euro-participation may result in sub-optimal short-run monetary management for individual countries. Furthermore, non-convergence for Greece and Spain implies that adopting the euro may have caused substantially asymmetric monetary shocks and macroeconomic imbalances. The experience of these countries suggests that for Latvia, Hungary, Poland and the UK, for which convergence was also rejected, adopting the euro in the foreseeable future may be significantly costlier than for the rest of the current “EMU outs”.

What are these findings’ implications in relation to the two questions earlier posed? First, convergence among the majority of EMU members suggests a common long-term effect of the ECB policy on national real interest rates. This meets a necessary (though not sufficient) condition for uniform long-run transmission of the single monetary policy, therefore it increases the probability of the ECB pursuing successfully its EMU-average price stability objective without causing large asymmetries in the output response of individual countries. From that point of view, given the relatively small (less than 20%)2 proportion of the high-growth, periphery countries (Cyprus, Greece, Ireland, Slovenia and Spain) in EMU’s total output; and assuming that (a) global inflation pressures persist in the short-run and (b) further shocks in global financial markets are avoided; it is more likely than not that in forthcoming quarters the ECB will increase its interest rate further to restore, with good chances of eventual success, euro-zone’s inflation back at its 2% objective.

If ECB nominal interest rates increase in the future, what are the chances of a recession following in the EMU high-growth economies? By having in recent years negative real interest rates, significantly out-of-sync with the EMU average, Greece and Spain appear to be more vulnerable that others to this risk: Over 1999-2007 nominal house prices increased by 187% in Spain and 75% (over 2001-2006) in Greece.3 During the same period, household borrowing increased significantly in both countries. Finally, starting from a balanced position in the mid-1990s, in 2007 the current account deficit reached the unprecedented values of 10% of GDP in Spain and 14% in Greece. These figures suggest significant macroeconomic imbalances. But having said so, crash-landing does not appear a foregone conclusion for these economies. A year since the onset of the credit crunch, economic growth (despite slowing down in recent quarters) remains strong in both countries; their house markets, despite having cooled significantly, have so far not shown signs of collapse; and their bank profitability remains healthy. Furthermore, both countries have some spare capacity for consumption smoothing, as Greece maintains a significantly lower household loans to GDP ratio compared to the EMU average (45.5% versus 60.2% in 2007)4 ; and Spain runs a public budget surplus (2.2% of GDP in 2007) providing it a significant margin for fiscal manoeuvre without violating the provisions of the Budget and Stability growth pact.

To sum-up, although Greek and Spanish authorities face country-specific risks not shared (at least to the same degree) by the rest of the EMU members, they have grounds to be optimistic that, in the absence of further global financial shocks, their economies will emerge from the current economic turbulence reasonably unharmed. If this turns to be the case, rather than causing a recession, further increases in ECB interest rates will trigger a soft- rather than a crash-landing and a reversion to a more sustainable path of equilibrium growth. Nevertheless, the risk of a significant economic downturn exists for both countries. This highlights the importance of eliminating real rigidities and promoting economic flexibility as a key mechanism of preventing fast growth causing unsustainable imbalances and increasing capacity to cope with unexpected external shocks. To ensure that the income catch-up process continues unabated, it is imperative for Greek and Spanish authorities to continue and intensify policies aiming to achieve structural reforms.

This has implications for the new EU-members aiming to join the euro in the foreseeable future. The Greek and Spanish experience suggests that for countries such as Latvia, Hungary and Poland pre-mature euro-accession is likely to cause the imbalances discussed above. The countries which have achieved real interest rate convergence are not immune to these risks either. Spain and Greece had done likewise prior to joining the euro, yet accession to the latter caused a downward RIRD structural break leading to the risks discussed above. The Greek and Spanish experience shows that pre-euro real interest rate convergence is a necessary but not sufficient condition for avoiding post-euro asymmetric monetary shocks; the latter critically depends on eliminating real rigidities and promoting economic flexibility. In the list of policy lessons to be drawn from last year’s turbulent events this certainly ranks among the ones at the top.


Footnotes:(1) E-mail: ArghyrouM@cardiff.ac.uk ; Web-page: http://www.cardiff.ac.uk/carbs/econ/arghyroum/ (2) Unless otherwise stated, the source of all quoted figures is Eurostat. (3) Figures taken from www.globalpropertyguide.com (4) Figure taken from the Hellenic Bank Association, www.hba.grReferencesArghyrou, M.G., Gregoriou, A., Kontonikas, A., 2008. Do real interest rates converge? Evidence from the European Union. Journal of International Financial Markets, Institutions & Money, doi: 10.1016/j.intfin.2008.05.004.

Clements B., Kontolemis, Z.G., Levy, J., 2001. Monetary policy under EMU: differences in the transmission mechanism? Working paper 01/102, International Monetary Fund.

European Central Bank, 2003. Inflation differentials in the euro area: potential causes and policy implications. European Central Bank: Frankfurt.