Abstract
This Selected Issues paper estimates the potential output and the associated nonaccelerating inflation rate of unemployment in the euro area. The study presents a conceptual framework for analyzing currency movements, and highlights the transmission of import price shocks on consumer prices. The paper compares different measures of trend money growth, and analyzes the monetary conditions. The study describes the stability and growth pact, outlines a simple framework for studying fiscal policy behavior, and estimates European Union countries' past cyclical fiscal policy responses to output growth fluctuations.
IV. The ECB’s Monetary Stance: First-Pillar Considerations1
A. Introduction and Summary
1. In October 1998 the European Central Bank (ECB) announced the Eurosystem’s “stability-oriented monetary policy strategy,” which would guide its monetary policy decisions in Stage 3 of European Monetary Union (EMU). Instead of announcing a target for money growth or inflation, as had been widely anticipated, the ECB chose instead to follow a strategy consisting of three main elements: a quantitative definition of the primary objective of monetary policy, namely price stability, and two pillars that would guide policy so as to achieve this objective.2
2. Under the first pillar a reference value for the growth of M3 is announced and revised on a yearly basis. This reference value is an important element of the monetary framework of the ECB since it allows the ECB to summarize the conditions prevailing in the money market: persistent deviations of nominal M3 growth from the reference value signal upside risks to medium-term price stability (Figure 1. top panel). Indeed, since January 1999, eight of the nine changes in the ECB’s main refinancing rate appear to have been in line with the assessment that first-pillar information warrants a change in the monetary stance (Table 1).
Euro Area: M3 Growth
Citation: IMF Staff Country Reports 2001, 201; 10.5089/9781451812992.002.A004
Source: European Central Bank and staff estimates.Euro Area: M3 Growth
Citation: IMF Staff Country Reports 2001, 201; 10.5089/9781451812992.002.A004
Source: European Central Bank and staff estimates.Euro Area: M3 Growth
Citation: IMF Staff Country Reports 2001, 201; 10.5089/9781451812992.002.A004
Source: European Central Bank and staff estimates.Euro Area: The ECB’s Monetary Policy Decisions
Staff interpretations based on various issues of ECB Monthly Bulletins.
Euro Area: The ECB’s Monetary Policy Decisions
| Main refinancing rate | Risks to medium-term price stability1/ | |||
|---|---|---|---|---|
| Date of policy decision | Level | Change | First-pillar assessment | Second-pillar assessment |
| 4/8/99 | 2.50 | -0.50 | neutral | downside |
| 11/4/99 | 3.00 | +0.50 | upside | neutral |
| 2/3/00 | 3.25 | +0.25 | upside | upside |
| 3/16/00 | 3.50 | +0.25 | upside | upside |
| 4/27/00 | 3.75 | +0.25 | upside | upside |
| 6/8/00 | 4.25 | +0.50 | upside | upside |
| 8/31/00 | 4.50 | +0.25 | upside | upside |
| 10/5/00 | 4.75 | +0.25 | upside | upside |
| 5/10/01 | 4.50 | -0.25 | neutral | upside |
| 8/30/01 | 4.25 | -0.25 | neutral | neutral |
| 9/17/01 | 3.75 | -0.50 | neutral | neutral |
Staff interpretations based on various issues of ECB Monthly Bulletins.
Euro Area: The ECB’s Monetary Policy Decisions
| Main refinancing rate | Risks to medium-term price stability1/ | |||
|---|---|---|---|---|
| Date of policy decision | Level | Change | First-pillar assessment | Second-pillar assessment |
| 4/8/99 | 2.50 | -0.50 | neutral | downside |
| 11/4/99 | 3.00 | +0.50 | upside | neutral |
| 2/3/00 | 3.25 | +0.25 | upside | upside |
| 3/16/00 | 3.50 | +0.25 | upside | upside |
| 4/27/00 | 3.75 | +0.25 | upside | upside |
| 6/8/00 | 4.25 | +0.50 | upside | upside |
| 8/31/00 | 4.50 | +0.25 | upside | upside |
| 10/5/00 | 4.75 | +0.25 | upside | upside |
| 5/10/01 | 4.50 | -0.25 | neutral | upside |
| 8/30/01 | 4.25 | -0.25 | neutral | neutral |
| 9/17/01 | 3.75 | -0.50 | neutral | neutral |
Staff interpretations based on various issues of ECB Monthly Bulletins.
3. Recently a number of studies have suggested that money has good leading indicator properties for inflation over the medium-term, including for the euro area, and can therefore play a useful role in the formulation of monetary policy (see Gerlach and Svensson, 2000; Trecroci and Vega, 2000, for example, and references therein). Making a judgment regarding the risks to price stability—stemming from monetary overhang—requires an estimation of an equilibrium, or trend, money growth for the euro area. The ECB’s reference value is one way of estimating the trend money growth of the euro area, using the quantity theory identity. However, a direct comparison of money growth with the reference monetary growth rate, albeit easy to understand and communicate, can be misleading. Instead, several indicators have been devised to sum up money market conditions—including the nominal and real money gaps and excess money demand—and a strategy has been launched by the ECB to publicize these to the wider public.
4. This paper derives and compares a number of different measures of trend money growth which are used to define indicators of “money gap” or “monetary overhang”; these are used to summarize information from the first pillar. In doing so, it also provides an in-depth analysis of monetary conditions in the euro area through the estimation of a money demand model. This analysis also throws some light on the source and magnitude of the observed money velocity trend.
5. All these different indicators are used to provide an assessment of the risks to price stability stemming from excess monetary growth. The main conclusions from this chapter are as follows:
Overall, the money market in the euro area is in broad equilibrium, thus signaling that medium-term inflation pressures are unlikely to arise from an accumulated “monetary overhang.” The results suggest that, although M3 has grown significantly over the past 18-24 months, this increase was, to a large extent, explained by developments in prices, GDP, and by interest rates;
The assumptions used in the derivation of money gap measures are crucial for the results and hence a careful analysis of the conditions in the money market is prudent. One key assumption, that is difficult to pin down accurately—due to uncertainties regarding its size and source—is the velocity trend for the euro area. It is shown that alternative models and assumptions regarding velocity trends can imply a range of reference values for M3 growth but overall these results suggest a reference value in the order of 5 percent—with a large band of uncertainty surrounding this estimate—compared with 4½ percent used by the ECB;
These uncertainties which are likely to become even more evident in the near future, during the changeover period of the euro, raise a number of questions regarding the ability of the ECB to effectively communicate information from the first pillar to the public. Hence, possible erratic movements in M3 growth over the next several months should be treated with caution.
B. Identifying A Reference Value for Money Growth
6. This section compares four methods for deriving a trend money growth rate or reference value for the euro-area M3. If money has any role, in terms of predictive power over future inflation, these trend money growth measures could be used to define useful indicators of excess money growth. Four different methods for identifying trend money growth are presented here: (i) the reference value for M3 growth announced by the ECB; (ii) a flexible velocity trend alternative to the ECB’s reference value as proposed by McCallum (1988, 1993); (iii) an estimate of trend money growth based on an unobserved component model; and, (iv) a measure of trend money growth based on a money demand model.
7. The results of applying these methods are presented in Figure 1 (middle panel) and a detailed explanation of the derivation follows in this section. The ECB’s reference value is compared with the flexible velocity trend alternative, the trend extracted using the unobserved component model, and finally the equilibrium money growth path obtained from the money demand model. The ECB’s reference value for M3 growth is estimated at about 4½ percent. Taking into account of velocity changes—using McCallum’s method—implies a reference value closer to 5 percent. This is corroborated by the estimation of the unobserved component model. Instead, the money demand models imply a wide range for M3 trend growth of about 3¾-5 percent based on a range of estimates of trend money velocity in the euro area.
Estimation using the quantity theory identity
8. Both the ECB’s methodology for estimating the reference value and McCallum’s formula use the quantity theory identity as a starting point and hence require assumptions regarding the trend growth for real output growth, prices, and velocity of money.3 The basic difference between the two methods is that, in deriving the reference value for M3 growth, the ECB assumes a constant velocity trend whereas McCallum’s formula estimates the velocity trend in terms of lagged (recent) values of money velocity; the advantage of the latter is that it puts more weight on more recent developments in money velocity and discounts past trends (both are compared in the middle panel of Figure 1).
9. The ECB’s reference value for nominal M3 growth is based on real potential output growth in the range of 2-2½ percent, inflation below 2 percent over the medium-run, and a constant rate of decline in M3 velocity in the range of ½-l percent. Taking mid-range values for potential output growth (2¼ percent), inflation (1½ percent), and velocity growth (-¾ percent), these values imply a reference value of 4½ percent.
10. The approach suggested by McCallum allows for a flexible velocity trend defined in terms of a moving average of lagged values of money velocity. This alternative reference value is calculated using the same values for output growth and inflation but a moving average of velocity over the 16 preceding quarters. According to this estimate the reference value appears to have increased following an acceleration in the (negative) velocity trend after 1999.
Estimation using an unobserved component model
11. Money is modeled as the sum of trend money growth, which is unobserved, and of an irregular component. An assumption is made regarding the specification of the trend process and such a model can be estimated with the Kalman filter.4 This “agnostic” approach to the measurement of the money gap allows us to freely estimate a (flexible) trend for money without the need to make any assumptions regarding potential GDP growth, long-run inflation, or velocity trend.
12. Estimation of the trend based on the unobserved component model requires knowledge of the model parameters. To simplify the analysis we impose specific parameters for the unobserved component model in such a way that the trend extracted with this method is equivalent to a the trend obtained through a Hodrick-Prescott (HP) filter.5 The results from this model also point to a higher trend growth compared with the ECB’s reference value of the order of 5 percent.
The money demand model
13. Finally, a long-run (real) money demand model for the euro area is used to measure the “desired” money demand stock at given levels of real GDP and the interest rate.6 The econometric analysis carried out for this paper shows that this relationship holds in the long run. Such a relationship suggests that—in steady state when the interest rate is unchanged—if prices and GDP are growing on a steady path consistent with potential GDP growth and the Central Bank’s inflation target, money demand should also grow steadily. In the long run nominal money supply should also grow at that same rate, although short-run deviations can persist for some time. Assuming that interest rates remain unchanged, velocity will only change proportionally to potential (or long-run) GDP growth if the income elasticity of money demand is different than one.7
14. A number of studies have confirmed the existence of stable money demand equations for the euro area (Coenen and Vega, 1999, Brand and Cassola, 2000, for example). Using these estimated (long-run) money demand equations it is possible to calculate the “desired” money stock which can be used, in conjunction with the actual money stock, to summarize the conditions prevailing in the money market. Nevertheless, a number of issues need to be addressed which can influence critically these estimates. One, which is important for the derivation of the reference value, is the measurement of the long-run velocity trend.
15. One of the stylized facts for the euro area is the constant velocity decline observed over the last 20 years. Different explanations for this trend can be advanced and these have different modeling implications:
Brand and Cassola (2000) argue that the finding of an income “elasticity” greater than unity accounts for the constant velocity decline in the euro area throughout the 1980s and 1990s. The reasons behind this could be the existence of wealth effects (i.e., money demand increasing faster than one would expect based on GDP growth). If that is true then in the absence of a proxy for wealth in the money demand equation, the scale variable also accounts for this “missing variable” and this is translated into an income elasticity greater than unity.
A second possibility is that the velocity trend is due to a steady (accelerating, perhaps, because of a world growth differential vis-à-vis the euro area) increase in the demand for euro M3 by foreigners.8 That may be difficult to model precisely, within a standard money demand model, although a deterministic trend could be used to capture the declining velocity trend given that such demand for M3 should be unrelated to euro-area GDP. If that is the case, an important question is whether that process could come to an end or whether the demand from abroad will continue to grow unabated; the introduction of euro notes and coins may—combined perhaps with the adoption of the euro as a reserve currency—contribute to such a development. A deterministic trend is not ideal for capturing the “true” data generating process given that demand for euro outside the euro area may fluctuate considerably. However, it can be used as a good proxy for world GDP, for example, and may provide a more reliable estimate for M3 demand in the event of a idiosyncratic downturn in the euro area.
On the other hand, the ECB admits that there are also good reasons to expect a reversal in this trend, for example, as a result of improvements in payment technologies and/or the disintermediation process taking place in the euro area.
Finally, the negative velocity trend in the euro area could be the result of declining inflation over the last 20 years. If that is the case, inflation could also be included in the money demand equation although the trend in nominal interest rates should explain part of the long disinflation process over the 1980s and 1990s and may be sufficient.
A vector autoregressive (VAR) model is constructed by including the following euro-area aggregates: the log of real money (M3), the log of real GDP, the short term interest rate (s), a long-term interest rate (1), and inflation defined as the (year-on-year) change in the log of the GDP deflator. This model includes a short- and long-term interest rate as it is found that such a specification provides a better model in terms of statistical properties. The data are basically the same as in Brand and Cassola (2000) but they have been extended to cover the period up to 2000Q4.
16. The (long-run) money demand equation estimated by Brand and Cassola (2000) and the alternative ones estimated in this paper are then used to calculate indicators for excess liquidity for the euro area.9 In particular:
The actual Brand and Cassola (2000) equation is given by: m-p=l .33y-l .60 Rl, where R1 stands for the long-term interest rate (henceforth denoted by BC model).
Our estimated version of the Brand and Cassola (2000) model yielded the following relationship: m-p=1.35y-0.6 R1, where R1 stands for the long-term interest rate (denoted by BC’).10
Finally, two alternative money demand models based on the models described above: m-p=1.04y+0.06 Rs − 0.47 R1 −0.001875t and m-p=y+0.29 Rs, −0.35 R1, − 0.95 Δ4p-0.001875t, where Rs stands for the short-term interest rate. The first allows for a deterministic trend while the second includes both a trend and inflation (denoted by “Md trend” and “Md trend, inflation”, respectively). In addition, these specifications allow for both the short- and long-term interest rate to affect long-run money demand.11
The last two specifications are preferred in terms of their statistical properties. These indicate that in the presence of a deterministic trend (and inflation), the income coefficient in the money demand equation is not statistically different from unity and that this is sufficient to describe the velocity trend in the euro area. Consequently, these results call into question the assertion that the declining money velocity is due to wealth effects—expressed in the model in terms of a high coefficient of GDP in the money demand equation.12 Instead, an exogenous decline in velocity—possibly due to demand for euros by foreigners—could also explain this trend. Allowing this (deterministic) trend to be freely estimated reveals a possible range for the velocity trend between -0.8 to -1 percent. This, also, would be consistent with a reference value closer to 5 percent compared with the 4½ percent adopted by the ECB.
C. Assessing Inflationary Risks Based on the First Pillar
17. To assess the conditions in the money market it is necessary to develop indicators that measure the “money gap” or “monetary overhang” in the economy. Two classes of summary measures of excess money liquidity are discussed here; a detailed discussion and comparison of these indicators is presented in Masuch, Pill, and Willeke (2001). For consistency purposes we adopt the terminology of the ECB regarding the names of these indicators. In particular, the “money gap” is simply the difference between money supply and “trend” money stock defined in terms of a quantity theory equation (i.e., assuming GDP growth in line with potential GDP growth and a stable long-run inflation rate—presumably consistent with the central bank’s medium term target; this was described in detail in the previous section.)13 On the other hand, the concept of “monetary overhang” compares the actual money stock with the estimated stock of money demand and measures the excess liquidity in the economy based on the difference between how much money people would like to hold (for a given, current, GDP and interest rates) and how much money is circulating in the economy.
18. Although there appear to be differences in terms of size, almost all indicators estimated here point to a closing of a (negative) gap by mid-2000 and. as a result of the decelerating money growth throughout the second half of 2000, a leveling of the money gap/overhang measures thereafter (Figure 1, lower panel). Assuming a slightly higher “equilibrium” inflation rate (of about 2½ percent) for the period 1994-97 would imply a higher trend growth of M3, and consequently a wider (more negative) money gap; this is also shown in the lower panel of Figure 1 (see reference value with break in inflation) for comparison purposes. Judging by these different measures, therefore, it appears that the money market is broadly in equilibrium and consequently it does not pose any risks to medium terms price stability.
19. Nonetheless, the derivation of these indicators is subject to a number of important caveats. In particular, to provide a measure of the money gap-using the ECB’s reference value and McCallum’s formula—a starting date, or base period, must be defined.14 Figure 2 (top panel) plots the money gap constructed using different base periods. The chart reveals that knowledge of the “correct” base period—which, in principle, should be chosen to be consistent with a period of money market equilibrium—is key in the measurement of the money gap. Second, the assumption regarding the velocity trend is also very important and the resulting differences under alternative hypotheses are significant and impinge upon the measurement of the money gap. Figure 2 (middle panel) plots the money gap under three alternative assumptions regarding the velocity trend: two that assume alternative negative velocity trends (-¾ of 1 percent and −1¾ percent annual percent change) and another one for unchanged velocity. In contrast, extracting the trend through an unobserved component model seems to be a good alternative since it is free of any assumptions regarding the base period or velocity.
Euro Area: Sensitivity of Money Gaps to Base Period, Velocity Trend, and Model Specification
Citation: IMF Staff Country Reports 2001, 201; 10.5089/9781451812992.002.A004
Source: Staff estimates.Euro Area: Sensitivity of Money Gaps to Base Period, Velocity Trend, and Model Specification
Citation: IMF Staff Country Reports 2001, 201; 10.5089/9781451812992.002.A004
Source: Staff estimates.Euro Area: Sensitivity of Money Gaps to Base Period, Velocity Trend, and Model Specification
Citation: IMF Staff Country Reports 2001, 201; 10.5089/9781451812992.002.A004
Source: Staff estimates.20. Figure 2 (lower panel) reveals differences among the monetary overhang indicators obtained using the money demand models described earlier. There appear to be significant differences between the indicators based on the preferred money demand specifications, reported in this paper, and the equations of Brand and Cassola. These differences are striking: a higher sensitivity of M3 to both the GDP and interest rates, in the Brand and Cassola model, explain the strong money demand since 1995 and hence the large (negative) monetary overhang in recent years—a period of declining interest rates and strong, uninterrupted GDP growth. This comparison highlights the importance of the money demand specification used in the derivation of the monetary overhang. Nevertheless, the preferred specifications seem to provide a reasonable (and stationary) measure for the euro area.
D. Conclusions and Policy Implications
21. The analysis presented in this chapter reveals that the money market is in broad equilibrium. Hence, medium-term inflation pressures are unlikely to arise from an accumulated “monetary overhang” in the euro area. Alternative summary measures of the conditions prevailing in the money market—including the money gap, and monetary overhang—are estimated and used in this analysis.
22. However, the derivation of these indicators is subject to a number of important caveats which are discussed in this chapter and hence a careful examination of the conditions prevailing in the money market is warranted. Such analysis, carried out in this chapter through the estimation of a money demand model provides evidence regarding periods of disequilibria in the money market and information about the likely source and magnitude of the velocity trend in the euro area. Overall, the results reveal that the trend money growth for M3 is estimated to be somewhere in the range of 4½ −5 percent, hence slightly higher than the reference value of the ECB. All of methods for estimating money trend for the euro area corroborate with this evidence. In addition, it is shown that there is considerable uncertainty surrounding the estimated velocity trend which is crucial in this analysis.
23. Looking ahead, the analysis and public communication of first-pillar information—to an already skeptical audience—are likely to become more difficult for two reasons. First, the stability of long-run money demand could be undermined by velocity shocks (this would increase the uncertainty about the size of money gaps and monetary overhang measures). Second, if the ECB succeeds in preserving a low-inflation environment, shocks to velocity—for example due to shifts in demand for euro notes and coins—relative to shifts in the money demand would dominate, making the signal extraction problem faced by the ECB even more difficult.
24. Although the finding of a stable money demand justifies the strategy of the ECB. the uncertainties pertaining to the velocity trend would call for a less precise reference value, perhaps in the form of a monitoring range, or uncertainty bands, for M3 growth. Such a strategy would allow the ECB to easily discount erratic fluctuations in M3 growth and would contribute to a more effective communication of its policy intentions in the future. In addition, the ECB should play down the importance of deviations of M3 growth from the reference value and should instead concentrate on the presentation of deviations of the stock of M3 from its trend level. A presentation of a wide range of indicators—complemented by analysis of M3 components and analysis based on money demand models—is desirable given the weaknesses of these measures.
References
Angeloni, I., Gaspar, V. and Tristani, O., (1999), “The Monetary Strategy of the ECB”, in D. Cobham and G. Zis (eds), From EMS to EMU: 1979-1999 and Beyond, New York: St. Martin’s Press.
Brand, C., and N. Cassola, (2000), “A Money Demand System for Euro Area M3”, ECB Working Paper No. 39.
Calza, A., D. Gerdesmeier, and J. Levy, (2001), “Euro Area Money Demand: Measuring the Opportunity Costs Appropriately”, forthcoming IMF Working Paper.
Coenen, G., and J.-L. Vega, (1999), “The Demand for M3 in the Euro Area”, ECB Working Paper no.6, www.ecb.int.
Doornik, J.A. and D.F. Hendry, (1994), “Modelling Linear Dynamic Econometric Systems”, Scottish Journal of Political Economy, Vol.41. 1–33.
Doornik, J.A. and D.F. Hendry, (1997), Modeling Dynamic Systems Using PC Fiml 9.0 for Windows, International Thompson Publishers.
European Central Bank, (1999), “The Stability-Oriented Monetary Policy Strategy of the Eurosystem”, ECB Monthly Bulletin, January, p. 39–50.
Gerlach, S. and L.E.O. Svensson, (2000), “Money and Inflation in the Euro Area: A Case for Monetary Indicators?,” http://www.iies.su.se/Ieosven/.
Harvey, A.C. and A. Jaeger, (1993), Detrending, Stylized Facts and the Business Cycle”, Journal of Applied Econometrics, 8, 216–237.
Hendry, D.F., (1995) Dynamic Econometrics, Oxford: Oxford University Press.
Hodrick, R. and E. Prescott, (1980), “Post-War U.S. Business Cycles: and Empirical Investigation”, Carnegie Mellon University, manuscript, reprinted in Journal of Money Credit and Banking, Vol. 29, No. 1, February, 1997.
Issing, O., V. Gaspar, I. Angeloni, and O. Tristani, (2001), Monetary Policy in the Euro Area, Cambridge University Press.
Johansen, S. (1988a), “Statistical Analysis of Cointegrating Vectors”, Journal of Economic Dynamics and Control 12: 231–54.
Johansen, S., (1988b), “Maximum Likelihood Estimation and Inference on Cointegration—with Applications to the Demand for Money”, Oxford Bulletin of Economics and Statistics 52: 169–210.
Johansen, S., (1992), “Testing Weak Exogeneity and the Order of Cointegration in UK Money Demand”, Journal of Policy Modelling 14: 313–34.
Johansen, S., (1995), Likelihood-Based Inference in Cointegrated Vector Autoregressive Models, Oxford: Oxford University Press.
Lutkepohl, H. (1994), “Interpretation of Cointegrating Relations-Comments on “Estimating Systems of Trending Variables”, Econometric Reviews, 13(3), 391–394.
Masuch, Klaus, H. Pill, and Caroline Willeke, (2001), “Framework and Tools of Monetary Analysis” in Monetary Analysis; Tools and Applications, European Central Bank, www.ecb.int.
McCallum, B.T., (1988), “Robustness Properties of a Monetary Policy Rule,” Carnegie-Rochester Conference Series on Public Policy, Vol. 29, pp. 173–204.
McCallum, B.T., (1993), “Specification and Analysis of a Monetary Policy Rule for Japan,” Bank of Japan Monetary and Economic Studies, November, pp. 1–45.
Trecroci, C. and J.-L. Vega, (2000), “The Information Content of M3 for Future Inflation,” ECB Working Paper No. 33.
Prepared by Zenon Kontolemis (ZKontolemis@imf.org). A longer version of this chapter containing technical details is available upon request from the author.
ECB (1999), p.46. For a useful justification of the ECB’s framework see Issing, Gaspar, Angeloni and Tristani (2001), and Angeloni, Gaspar, and Tristani (1999), for example.
In logs the rate of change of money growth can be expressed in terms of the rate of growth of potential GDP, long-run inflation and velocity, or, Δm* = Δy* + Δp* - Δv*.
Specifically,
For a given set of parameter estimates (δ1=2. δ2=-l and var (ξ)/var (ε)=l600) this model is equivalent to the HP filter (see Hodrick and Prescott, 1980, and Harvey and Jaeger, 1993, for example).
A standard money demand equations in logs has the form, md - p = α0 y -α1 R, where p is the (log) GDP deflator, y is (log) GDP, and R is a measure for the opportunity cost of holding money.
By definition velocity is given by, v = p + y - m = (1 - α0)y - αtR, which in growth rates can be written as ẏ = (1 − α0)ẏ − α1Ṙ. However, assuming that interest rates remain unchanged, velocity will change proportionally to potential (or long-run) GDP growth if α0 ≠ 1.
According to estimates by the Deutsche Bundesbank, in the mid 1990s, 30-40 percent of all the DM banknotes were held abroad. Although, estimates for other currencies are not available it is certain that smaller quantities of other currencies may also be circulating outside the euro area (see Box 1, of the ECB’s Monthly Bulletin, September, 2001, for a discussion).
The long-run equations are identified using the Johansen procedure; see. for example. Johansen (1988a, b and 1995), Hendry (1995), Doornik and Hendry (1997), and references therein.
The VAR model estimated by Brand and Cassola (2000) includes two lags of each variable. In contrast, the results reported here are based on a model estimated with five lags which are all found to be essential to provide a well-specified model.
A discussion about the correct measure for the opportunity cost of holding broad money demand for the euro area can be found in Calza, Gerdesmeier and Levy (2001).
The word elasticity is not used here since it is argued that the parameters in these equations can not, strictly speaking, be interpreted as elasticities (i.e., showing the responsiveness of one variable to a change in another keeping all other variables unchanged); this point is explained in more detail in the longer version of the paper.
Other indicators, including the real money gap, are variations of the money gap and the discussion that follows also applies to these concepts.
