BETTER OFF WITHOUT THE EURO? EVALUATING MONETARY POLICY AND MACROECONOMIC PERFORMANCE FOR THE UK. AND SWEDEN.
| Autor | Krause Montalbert, Stefan |
| Cargo | Ensayo |
| Páginas | 133(19) |
RESUMEN
En este artículo comparo los resultados macroeconômicos, medidos por la volatilidad del PIB real y la inflación, para dos de los países que optaron por no entrar en la UEM: el Reino Unido y Suecia. En particular, estoy interesado en saber cuánto de los cambios en el desempeño macroeconômicos experimentados por el Reino Unido y Suecia después de 1999 se debe al aumento de la eficiencia en la conducción de la política monetaria independiente. El objetivo es analizar los cambios en la volatilidad macroeconômica que se podrían haber alcanzado si estos dos países hubieran adoptado el Euro a partir de enero de 1999.
PALABRAS CLAVE: EUROZONA, POLÍTICA MONETARIA, VOLATILIDAD DE LA INFLACIÓN Y DEL PRODUCTO.
ABSTRACT
I compare macroeconomic performance, measured by the volatility of real GDP and inflation, for two of the countries that opted not to enter the EMU: the United Kingdom and Sweden. In particular, I am interested in finding out how much of the macroeconomic performance changes experienced by the U.K. and Sweden after 1999 is due to increased efficiency in the conduct of independent monetary policy. Eventually, the objective is to analyze whether or not further changes in macroeconomic volatility could have been attained if these two countries would have instead adopted the Euro starting January 1999.
KEYWORDS: EURO AREA; MONETARY POLICY; INFLATION AND OUTPUT VOLATILITY.
-
INTRODUCTION
In this paper, I compare macroeconomic performance, measured by the volatility of real GDP and inflation, for two of the countries that opted out of joining the European Economic and Monetary Union (EMU) in 1999: the United Kingdom and Sweden. In particular, I am interested in finding out how much of the macroeconomic performance changes experienced by the U.K. and Sweden after 1999 is due to increased efficiency in the conduct of independent monetary policy, as opposed to changes in the impact of demand and supply disturbances. Eventually, the objective is to analyze whether or not further changes in macroeconomic volatility could have been observed if these two countries would have instead adopted the Euro starting January 1999.
The study's objective is to shed some additional light on analyzing the pros and cons of EMU membership; while revisiting some past (albeit once again relevant) questions, including:
* Are changes in fluctuations due to policy, shocks, structural changes, or other factors (Ahmed, Levin, & Wilson, 2002; Dynan, Elmendorf, & Sichel, 2006; Herrera & Pesavento, 2005; McConnell & Perez-Quiros, 2000; Stock & Watson, 2002, 2003)?
* Are Inflation Targeting (Sweden and the UK), and price stability as the main mandate (EMU) still the best policies for minimizing fluctuations (Bernanke, Laubach, Mishkin, & Posen, 1999; Krause & Méndez, 2008; Mishkin & Schmidt-Hebbel, 2002; Rudebusch & Svensson, 1999, 2000; Walsh, 1995)?
* Do Sweden and the UK have anything to gain at all if they were to join the EMU (Alesina & Barro, 2002; Pesaran, Smith, & Smith, 2005; Rose & Engel, 2002; Sõderstrõm, 2008)?
Figures 1.1 and 1.2 document the changes in macroeconomic volatility that the U.K. and Sweden have experienced between the pre- and the post-Euro periods, the latter of which is again divided into the subperiods before and after Lehman Brothers' filing for Chapter 11 bankruptcy on September 2008, and the ensuing financial crisis and worldwide recession. For comparison purposes, I also include the data for the Euro Area as a whole; using a weighted average for the period comprising 1991 and 1998. Given the quarterly data availability, I will focus on the following three subperiods throughout the paper:
* Period 1:1991:QI-1998:QIV
* Period 2: 1999:QI-2008:QII
* Period3: 2008:QIII-2010:QIV
The top part of Figure 1.1 reports the standard deviation of the real GDP output gap (measured as deviations from an HP-trend with a parameter of 1600); while the bottom part reports the standard deviation from the actual growth rate data series. For the U.K., real volatility either remained unchanged or slightly increased when comparing the 1991:QI-1998:QIV and the 1999:QI-2008:QII periods, depending on whether I consider the volatility of the output gap or of GDP growth, respectively. In the case of Sweden, both measures of real volatility suggest a decline between Period 1 and Period 2. Finally, both countries experienced a substantially larger variance after 2008:QIII, when compared to either of the two prior periods.
FIGURE 1.1: CHANGES IN REAL GDP VOLATILITY (1991:I-1998 IV / 1999:I-2008:II / 2008:III-2010:IV)
Figure 1.2 shows how the standard deviation of inflation has changed between the three periods, with the top part reporting the volatility measure centering average inflation at a 2% level for all 3 periods, and the bottom part reporting it using each period's average inflation as the center point. Given the relatively low levels of inflation prevalent for both countries since the beginning of the 1990s, the magnitudes of the standard deviations are very similar. The general observation is that for Sweden and the U.K., inflation volatility fell by about 30%-35% when comparing Period 1 and Period 2, and then in Period 3 it returned to near its value of Period 1.
FIGURE 1.2
CHANGES IN INFLATION VOLATILITY (1991:I-1998 IV / 1999:I-2008:II / 2008:III-2010:IV)
The above figures serve not only to show the actual changes in macroeconomic performance that both countries experienced over the past 20 years, but it also helps to illustrate that the measured changes in the volatility of both inflation and real output do not change perceptibly with a different choice of center points or targets. I shall return to this last point later in the paper.
The remainder of the paper is organized as follows. Section II details the set-up and estimation of an aggregate demand - aggregate supply (AD-AS) model, that will later be employed to assess the role monetary policy has played in the observed changes in macroeconomic performance in Section III. It also lays out the groundwork for the counter-factual analysis used to gauge macroeconomic performance under the Euro. Finally, Section IV presents some concluding remarks.
-
METHODOLOGY
Estimating the AD-AS model
Estimating the dynamics of the output gap and inflation plays a dual role in assisting us to perform the task at hand: First, the estimated coefficients are used to derive the Taylor (1979) frontier for each country and each subperiod; this efficiency frontier is the main identification tool I employ to determine how much of the observed changes in macroeconomic performance are due to monetary policy, as in Cecchetti, Flores-Lagunes, and Krause (2006). I describe this in more detail on Section III. Second, the specification will serve to perform the counter-factual analysis, and generate alternative scenarios for the dynamic behavior of real GDP and inflation under the adoption of the Euro, as explained on Section IV.
The AD-AS model should be general enough for the countries of interest, to allow for direct comparisons; and meet all model specification criteria. With this in mind, I estimate the following specification:
(y)t = [[alpha].sub.10] + [2.summation over ()l=1] [[alpha].sub.1l][y.sub.t-l] + [phi](lt-1 - [[pi].sub.t-1]) + [[psi].sub.y] [[pi].sup.c.sub.t-1]) + [U.sub.yt], (1)
[[pi].sub.t] = [[alpha].sub.20] + [4.summation over (l=1)][[alpha].sub.2l][[pi].sub.t-l] + [[psi].sub.[pi]] [[pi].sup.x.sub.xt-l] + [U.sub.[pi]t]. (2)
This formulation is based on the empirical observation that monetary policy actions affect output before inflation (see, for instance, the theoretical model of Svensson, 1997; and the empirical model in Rudebusch and Svensson, 1999; among others). The error terms assumed to be uncorrected, and to have zero mean and constant variance.
The first equation represents an aggregate demand or IS curve. It relates the output gap y to two of its own lags to account for the persistent nature of real GDP (2); and one lag of the real interest rate i-[pi], which captures the effect of intertemporal substitution in consumption (Clarida, Gali, & Gertler, 1999, 2000). The second equation is an aggregate supply or Phillips curve. Here, inflation is assumed to be a function of four of its own lags, and two lags of the output gap accounting for the pressure of increased economic activity on prices.
When dealing with open economies, it's important to gauge the effect of external shocks. The option employed in this study is to augment both equations with Euro Area inflation translated into domestic currency [[pi].sub.x] (lagged one period). This is done to take into consideration the inter-relation between the economy of interest and that of Euro Area(3).
I estimate equations (1) and (2) for each country separately via OLS for the pre- and post-Euro years, using quarterly data taken from Datastream (4).
Table 1.1 reports the estimates of the output gap equation (1) for the U.K. and Sweden for two subperiods. Period 1 as defined on Section I, includes the pre-Euro years (1990:QI-1998:QIV). Given the relatively small size of Period 3...
Para continuar leyendo
Solicita tu prueba