Endogenity of the monetary transmission mechanism

Document Sample
Endogenity of the monetary transmission mechanism Powered By Docstoc

                                       BÁLINT HERCZEG
         University of Debrecen, Faculty of Economics and Business Administration

                                first version: 12th January 2009

                  Preliminary results, work in progress, please do not quote.

The aim of this paper is to show that the monetary policy transmission mechanism is an
endogenous condition of an optimal currency area, because the integration of the financial
system changes makes the transmission mechanism more.
Keywords: monetary transmission, optimal currency area, VAR
Journal of Economic Literature (JEL) code: …

1. Introduction
There is a large part of the economic literature that tried to predict the effects of the European
Monetary Union on the member states. One base for these papers is the so called Optimal
Currency Area (henceforth OCA) literature which originates from Mundell [1961]. According
this literature one of the primary conditions of the desirability of common currency, so that
the advantages of the common currency (mostly on micro level) outnumber the disadvantages
of the giving up own monetary policy, is the synchronization of the business cycles. In this
case the timing and size of common monetary policy moves will be favourable for every
participant. Frankel and Rose [1996] showed that this condition should be endogen, because
the common currency busts trade among the participants and the trade connections will work
for the synchronization of the business cycles.
The role of monetary transmission mechanisms (henceforth MTMs) in the OCA literature is
not so strait forward. It is not told anywhere in the OCA literature that the same monetary
policy move should have the same effect in size and timing in the countries using the common
currency. But the similarity of MTM should serve as a condition and there were many papers
which tried to measure the differences of the transmission in the EMU, because "(i)f the
timing and intensity of the effects which a European Central Bank (ECB) action can trigger

will differ between the participants, asymmetric shocks will be the outcome of EMU."
(Ehrmann [1998] p. 4.).
The question I would try to answer in this paper is whether the common currency changes the
MTM of the different countries, making them more similar or to use the term used by Frankel
and Rose [1996] is the MTM endogen1 in the EMU?


The paper organized as follows: first I briefly show the stages of MTM and the different
MTM channels, then I summarize the results of the previous papers and articles, that tried to
investigate the differences and similarities of the transmission in different EMU countries.
Using the methodological lessons learnt from this section, I first analyse the variables that
indicate that the integration of the financial system diminishes the differences in the
transmission, then I run small vector autoregressive model, to see if this process can be seen
in the IRF-s. The last section concludes.

2. Transmission of the monetary policy
As a first step the decision making board of the monetary policy sets the value of the
instrument according to its aim (some definition of price stability). In an inflation targeting
(henceforth IT) framework this would mean that if the forecasted inflation diverges from the
aim, the board would increase the instrument interest rate. The instrument variable is
changing the prices and liquidity of the other assets in the financial system (through direct
channels or through influencing the expectations of the agents). These changes in the financial
environment influence the decisions and choices of the household and the business sector – so
it influences the demand side of the economy. The altered demand side has it effects on the
inflation dynamics.
So the schema to describe these relationships is:

Figure 1; Schema of the monetary transmission
aim  instrument variable  price of other financial assets  behaviour of households
                      and firms (aggregate demand)  inflation dynamics

  The term endogeneity in connection with the monetary transmission mechanism was first used to my
knowledge by Arnold and de Vries [1999]

Several channels of monetary policy transmission are distinguished in the literature according
to which assets' prices the monetary policy's instrument variable changes and how it affects
the behaviour of the agents of the economy2:
- exchange rate channel has its influence through the changes in foreign demand and supply
or through wealth channels (foreign currency denominated debt)
- through interest rate channel monetary policy can influence the opportunity cost/cost of
capital and thus investments and consumption (substitution effect) as in Taylor [1995] "A
slightly more specific is to say that the interest rate channel (IRC) is the mechanism that
operates in the absence of capital market imperfections, i.e. only through intertemporal
reallocation of expenditures that follows a change in expected real interest rates." Angeloni
and Ehrmann [2003b] p. 15.
- in case wealth channels3 interest rate is considered as a discount factor which can alter the
value of assets, so it can have its influence on households directly through their wealth
(wealth effect) or through the changes of their interest rate income and so their disposable
income (income effect). The same can be true for enterprises4.
- credit channel is complementary to the previous channels inasmuch the effect of a shock
created by the monetary policy amplifies the problems caused by asymmetric information in
the financial system. This can prevent firms from investing and households from smoothing
their consumption, as the financial assets value could decline and so not be able to play the
role of collateral (balance sheet effect). Or the frictions in the financial system can prevent
banks from getting refinance loans from the central bank and through it forcing them to
decrease their lending (bank lending channel). For the bank lending channel to work one
would need small banks that depend on central bank finance, and firms that can not find a
substitute for bank loans at the capital market.5 This also means that through this channel the
monetary policy has allocative effects as well, because after an interest rate increase not worst
investments are cancelled, but the investments planned by smaller firms, which could have
social costs Cecchetti [1995].
- in case of expectation channel, the monetary policy's credible communication can affect the
agents behaviour through affecting their expectations and so their economic decisions about
the future.

  See for example Mishkin [1996] for a more detailed review
  In an alternative grouping of the transmission channels the interest rate channel is divided up in 3 subchannels:
substitution effect, wealth effect and the income effect.
  See Tobin [1969] for the Tobin's Q effect.
  See Bernanke and Gertler [1995] for classic review or Bernanke [2007] for a new literature.

In the next sections I will look through what differences were found in the different stages and
channels of the monetary transmission, and what changes were caused by the common
currency and common monetary policy in these channels.

3. Were there differences before the EMU?

         "The existing empirical evidence does not give a clear picture of how important these
           differences actually are, or if they even exist." (Elbourne and De Haan [2004] p. 1.)

In this section I will summaries the different empirical papers that reason alongside or against
the "dangerous" differences of the monetary transmission mechanism, showing the source of
puzzle behind the quote above. I summarize the parts of these papers that concentrate on the
state of transmission deducted from data before or shortly after the begin of the third stage of
the EMU. If one tries to draw conclusion from data before a regime change, one must be
cautious, because "(a)t least since Luces' celebrated "critique", it has been recognized that
changes impact on the expectations formation mechanisms and hence on the economy's
response t policy. It is hard to think of examples where this line of reasoning could apply
more forcefully than EMU." Angeloni and Ehrmann [2003b] p. 6.

Various methods were used by the authors, so the papers are grouped according their

3.1. Key variables
This method is most widely used for two reasons, first the collected key variables are
supposed to be in connection with one of the channels of transmission above (see table 1.) and
this way the authors try to catch the strength of the different channels. The second reason
expressed by Guiso et al. [2000], is that they believe that this key variables refer to real
structural relationships (behavioural responses), which ought to their longer accommodation
time could survive a regime change, so expose more about the true relationships that could
characterize the transmission of the ECB than the other methods (VARs etc.). According
Suardi [2001] the problem with this method is that the strength of a channel can only be
compared relative the same channel in another country and nothing can be told about the
different importance of channels inside a given country. This way the method can't show the

differences in the strength of the complete transmission. Guiso et al. [2000] looks at this as an
advantage, because they intended to look at the channels to avoid the aggregation bias.

Table 1; Indicators of strength of different channels of monetary transmission could be
measured with.
1, Interest rate pass through (instrument variable  price of other financial assets)
           competition among banks (HM Treasury [2003], Mojon [2000])
           weight of loans with variable interest rates (HM Treasury [2003])

           monetary regime (inflation, volatility of money market rate) (Mojon [2000])
2, price of other financial assets  behaviour of households and firms (aggregate demand)
interest rate channel
           sectoral composition of the economy (Dedola and Lippi [2005], HM Treasury [2003])
           capital output ratio (investment/GDP) (Guiso et al. [2000])
wealth channels
           structure of households' and firms' asset and liabilities (interest rate and term) (HM Treasury [2003],
           (Guiso et al. [2000]))
           fraction of financing that is short term (Guiso et al. [2000])
           household indebtedness (Guiso et al. [2000])
           firm's median leverage (Guiso et al. [2000])
exchange rate channel
           exchange rate pass through (HM Treasury [2003])
           weight of foreign currency denominated assets (HM Treasury [2003])
           openness to trade (net export outside EMU) (Guiso et al. [2000], Suardi [2001], Peersman [2001])
credit channel
           size structure of firms
           median number of employees per firm (Guiso et al. [2000])
           the debt structure of SMEs (HM Treasury [2003])
           possibility of alternative finance
           ratio of firms with single bank (Guiso et al. [2000])
           use of collateral (Suardi [2001])
           market capitalization relative to GDP (Guiso et al. [2000], Kashyap and Stein [1997])
           health of banking system
           average bank size (Guiso et al. [2000])
           ratio of total deposits in the 5 largest bank (Guiso et al. [2000], Kashyap and Stein [1997])
           efficiency of legal system and contract enforcement (measure of asymmetric information)
           months to repossess (Guiso et al. [2000])
           repossession cost as a % of house value (Guiso et al. [2000])
3, behaviour of households and firms (aggregate demand)  inflation dynamics
           employment protection (OECD index) (Guiso et al. [2000], Suardi [2001])

3.1.1. Interest rate pass-through
Mojon [2000] finds that the response of short-term credit rates to money market rates is
significantly smaller in Italy, Germany and Spain, than in Belgium, France and the
Netherlands. He also finds that the financial system influences the pass through in which the
monetary regime (inflation +, volatility of money market rate -), the competition among banks
(+), the rigidity of founding costs (higher staff cost -) have significant effect on the speed and
proportion the banks' credit and deposit interest rates adapt to the money market rates.

3.1.2. Interest rate and aggregated demand

Exchange rate channel
Suardi [2001] finds that exports to third countries are proportionally larger in the small open
economies, 30% in Belgium and 22% in the Netherlands. In Germany it is 17% of the GDP,
13-14% in France and Italy, 11% in Spain according the 1999 estimates of the EU
Commission. On the other hand as the import also tends to be bigger in the small open
economies (30 and 32% in Belgium and the Netherlands, 15% in Germany, 12% in Spain and
France and 11% in Italy), the differences above don't indicate that the differences in the
exchange rate channel should be large. It is interesting to see that the same time bigger states
tend to export larger share of their export outside the EU 15, as can be seen in the forth row in
Guiso et al. [2000]'s table 3.

                          Table 2; Measures of exchange rate channel.
                                 Extra    EU15       Extra euro area      Extra euro area
                                 export / GDP        export of goods      import of good
                                 (1996) a            and services (% of   and service (%
                                                     GDP) b               of GDP)b
                 Germany              10.4                   17                 15
                 Netherlands          15.6                   22                 32
                 Belgium              19.9                   30                 30
                 France                8.9                   13                 12
                 Spain                 7.1                   11                 12
                 Italy                12.2                   14                 11
                 Austria              17.6                    -                  -
                 Finland              18.1                    -                  -
                 Ireland              24.4                    -                  -
                 Portugal              6.0                    -                  -
                   Peersman [2001] p. 22., bSuardi [2001] p. 28.

Interest rate channel
The structure of production can shed light on differences, on the one hand sectors with
interest rate sensitive demand could be affected by changes of the monetary policy (mainly
sectors producing durable goods), on the other hand capital intensive sectors (mainly heavy
industries, like machinery and transport) can be affected by changes of the alternative costs of
capital. Dedola and Lippi [2005] finds that Germany is more specialised in durable goods
production than Italy, France and the U.K. And this way it is more like to be effected harder
by an interest rate shock. The level of investment tends to be higher in the six euro area
countries on the average in 1999-2000 (19-23% of the GDP), than in Sweden and or the U.K.
(16-17%) according EU Commission's estimations, that's why Suardi [2001] thinks that Italy

and Germany should be more sensitive. Guiso et al. [2000] uses the ratio of fixed capital to
output to reveal the need for investment. According to Guiso et al. [2000] the higher this ratio
is, the more investment is needed to keep this ratio up, the more investment is needed the
more interest rate sensitive it should be. As can be seen in second row in table 3., here again
Germany should be more sensitive. Guiso et al. [2000] also uses the maturity structure to
show the exposure of firms to interest rate changes, the investment needs of the German firms
is balanced by the small fraction of short term, interest rate sensitive financing. Also the
Netherlands have low fraction of short term financing.

Wealth and income channel
Suardi [2001] shows that by 1998 data most countries negative net interest bearing assets
(except Belgium and Italy), and household dept to disposal income is the largest in the
Netherlands (154%) and the smallest in Italy (37%). In Italy and France the interest rate of the
liabilities are almost entirely fixed, in contrast with UK, Sweden and Spain where the interest
payment is linked to a short-term market rate, in Belgium and Germany renegotiable rate
mortgages are widespread. This suggest that the income effect should be stronger in UK and
Sweden, Italy (because less liability, and big stock of government bonds).
Guiso et al. [2000] Italy is with the lowest borrowing level, although Belgium also shows
quite low levels, according to them in this countries the households couldn't borrow to smooth
their consumption, credit rationing.

Table 3; Selected characteristics for European countries
Variable                                                              Countries
                                        UK      Germany     Italy    France     Spain   Netherlands   Belgium
employment protection                   0,9         2,6       3,4       2,8       3,1        2,2         2,5
(OECD, rank from 26)                    (2)        (20)      (23)      (21)      (22)       (13)        (16)
capital output ratio                   1,99         4,0       3,2       3,0      NA.       N.A.          3,0
(investment/GDP)                      (0,154)    (0,223)   (0,180)   (0,191) (0,212)      (0,197)     (0,181)
fraction of financing that is short
                                      0,960      0,593     0,838     0,893    0,925       0,620        0,882
exports outside EU-15 relative to
                                       0,47       0,44      0,45      0,38     0,29        0,25        0,29
                                       63,1       52,0      52,3      46,3     53,5        43,9         51,4
firm's leverage (median) %
                                      (60,5)     (61,0)    (62,5)    (49,1)   (56,4)      (63,7)       (58,4)
median number of employees per
                                      1,128       406       251       357      267         205          363
household indebtedness                1,020      0,779     0,314     0,510    0,580       0,649        0,415
months to repossess                     12        15         48        11       36         2,5           24
repossession cost as a % of house
                                       4,75        6         19        15       10          11         19,5
% of firms with single bank            22,5       14,5       2,9       4        1,5        14,3         0
market capitalization relative to      1,65       0,48      0,46      0,65     0,69        1,53        0,94

average bank size billions of dollar    24,9   12,8   12,3   20,1    10,2      32,1       22,3
% of total deposits in 5 largest bank   27,0   14,0   40,4   68,8    39,8      81,3       61,0
source: Guiso et al. [2000]

Credit channel
Guiso et al. [2000] uses number of months needed to repossess collateral and the legal cost of
repossession as a measures of strength of the financial market institutions. Italy, Spain and
Belgium show long periods to repossess a house, with high repossession costs as well, on the
other end of the spectrum is UK together with the Netherlands and Germany.

To measure the importance of small banks Kashyap and Stein [1997] uses the ratio of assets
controlled by 3, 5, 10 largest institutions/commercial banks, and they point to a similar
picture: there tend to be larger banks in dominant position in Belgium and Netherlands, there
is a smaller concentration in/smaller banks control significant fraction of the assets in Italy,
Germany and Luxembourg.
As "healthier" banks can more easily find alternative financing in case of a monetary
tightening Kashyap and Stein [1997] uses ROA calculated from the OECD statistics to proxy
for bank health. According to this measure the banks in Netherlands and Luxemburg are in
good shape, whereas the banks in France and Italy appear to be weak, with high levels of bad
loans and low profit rates.
Smaller firm depend more on bank financing because the monitoring cost would be to high to
get alternative financing from the capital markets. Kashyap and Stein [1997] looks at the
distribution of workforce between different firm sizes, and finds that in Spain and Italy more
than 40% of the workforce work for firms with fewer than 10 people. Small firms seems to be
relatively important employer also in Greece, Netherlands, Portugal. In Germany,
Luxembourg are dominated by larger firms.
The chance of alternative financing depends on the development of the capital markets
proxied by Kashyap and Stein [1997] with the size of these markets. The availability of non-
bank finance is greatest in Belgium and Denmark, conversely Greece, Italy and Portugal
appear to be the least developed by this metric.
The overall importance of the credit channel seems to be high in Italy and Portugal (smaller
unhealthy banks, bank dependent small firms), and low in Belgium and the Netherlands
(large, healthy banks and large firms, with opportunity of capital market financing). The

picture is somewhat mixed in case of Germany, Luxembourg (small but healthy banks, larger
firms), and average in case of Denmark, France, Ireland, Greece and Spain.

Suardi [2001] is also uses almost the same indicators with result Italy and Spain are more
exposed to the balance sheet effect because greater number of smaller firms and a
comparatively weaker legal system. A new variable to measure the effect of balance sheet
channel is the use of collateral, data from 1993 shows that the percent of total loans to non-
bank private sector backed by real estate collateral is between 33%(Spain)-41%(France),
against the 59% in UK and more than 61% in Sweden.

Cecchetti [1999] also uses the lending channel to show the asymmetries in the MTMs among
the EMU countries. His analysis starts at institutional level, because he explains the
differences in financial structure with the differences in legal structure (categorized by the
lines of shareholder rights, creditor rights, enforcement). According his argument the
differences caused by the different financial structures (so the different strength of the credit
channels) ought to survive the regime change because they root in the differences of the legal
systems, and the changes at the institutional level of the economy are very slowly. "The
arguments presented here suggest that unless legal structures are harmonized across Europe,
financial structures will remain diverse, and so will monetary transmission mechanism. It will
not be enough to make regulatory structures more similar, since such a change will not, in
and of itself alter the structure of capital markets." Cecchetti [1999] p. 22. He compares
different countries and comes to the conclusion, that the credit channel should be the most
effective in Italy and Austria, and the least effective in Belgium, Ireland and the Netherlands.
Elbourne and De Haan [2004] examined Cecchetti [1999] quantitative results, as he tries to
verify that his VAR analyses are in positive correlation with the indexes made of the key
variables. Elbourne and De Haan [2004] founds that the correlation is often negative if one
looks at the components of indexes, so "that the result of Cecchetti may be result of
aggrageting too far" Elbourne and De Haan [2004] p. 21.

3.1.3. Aggregated demand and the inflation
As a measure of the sticky prices the OECD's strictness of employment protection legislation
is a popular ranking used by several authors (for example Suardi [2001], Guiso et al. [2000]).
According this measure the job protection is lower in U.K. than in any other country, and

fairly similar in the other countries. Slightly weaker job protection can be found in the

3.1.4. Overall
Using this methodology Suardi [2001] came to the conclusion that the structural differences
across the six euro area countries (Belgium, Germany, Spain, France, Italy, the Netherlands)
are of lesser scale than those between them and the UK or Sweden. The stress is on difference
and not on in which countries it will be stronger.
Guiso et al. [2000] predicts in their paper that the Italian economy appears to be one in which
several of the theories would predict a strong effect on monetary policy on the economy
(fairly high fixed-capital ratio, poor contractual enforcement, lots of small firms, rigid labour
markets and small banks operating in a bank dominated financial system). UK is the opposite,
Germany …

Figure 2; Overview of the channels of monetary transmission in Europe.

Peersman [2001] p. 33.

3.2. Models
3.2.1. Vector Autoregressive Models

Sims [1992] was the first, who tried to explain the problem of price puzzle (positive initial
reaction of prices on an interest rate increase) and propose a treat. According his explanation
the problem is caused by the fact, that the monetary policy uses a wider informational set,
than the one usually built into the models. So as the monetary policy sees inflation pressure to
build up, might raise the interest rate in advance and so the increased interest rate and the
higher prices happen in the same time. In this case the monetary policy isn't identified
appropriately because the reaction of an unobserved variable isn't an exogenous shock to the
variables. The proposed solution could be to widen the informational set in the model, to filter
out what is the monetary policy react to. Sims proposed a commodity price index for this task.
His result for France and Germany this identification is helped to reduce the price puzzle,
even if it doesn't managed to eliminate it. Despite the price puzzle the reduction of the
industrial production is in both cases considerable (not significant because Sims [1992] didn't
estimate confidence intervals), in Germany almost twice as high as in France (if this statement
makes sense anyway in case the monetary policy isn't properly identified as an exogenous
monetary policy action – even that result could be questioned, which leads back to argument
over VAR see for example Cochrane [1998], Rudebush [1998], Sims [1998])

Kim [1999] used a five variable SVAR, next to more common variables like call money rate
(interest rate), monetary aggregate, consumer price index and industrial production, he used a
the world export commodity price index (CMPW) in terms of domestic currency to capture
foreign inflationary shocks and exchange rate movements that can be important in smaller
(compared to the USA) economies. Because the identification scheme is very similar to that
of Kim and Roubini [2000], here again the monetary policy is identified through its reaction to
changes of monetary aggregate and the CMPW. Kim [1999] finds that monetary policy, in all
G7 countries, unexpectedly increase interest rate in case of positive shock of monetary
aggregate or world export commodity price index, taking contractionary positions. The
impulse response functions behave in accordance to the expectation in sign and duration, with
significant reactions (on exception is the reaction of the German CPI to an interest rate shock,
which although declines steadily, has a very wide confidence interval including the 0
reactions as well). The IP response is almost identical among the three EMU countries that in
the sample.

Kim and Roubini [2000] main aim was to get rid of the different puzzles, so they don't
compare their IRFs of different countries. They use a blockwise identification in their 7

variable SVAR, with information variables like oil prices and the US FFR, and because they
run this specification on open economies (Germany, France, Italy), they have to build in the
exchange rate. The identification of the monetary policy is that it, controlling the interest rate,
has an immediate reaction on, monetary aggregate, oil price and exchange rate changes. These
coefficients are mostly negative which implies that the monetary policy increases interest
rates when it observes unexpected increase in the monetary aggregate and in the price of oil,
and unexpected exchange rate depreciation. So monetary policy is leaning against the wind
and takes contractionary positions. One exception is the reaction of Italian monetary policy on
oil shocks. The IRFs show that the magnitude and persistence of the interest increase and the
money supply decline differs among countries, but in all of them the effects are statistically
significant on impact and over the medium run. In Germany and Italy the fall in price level is
persistent and significant over the full 48 months horizon, in France the price level
insignificantly increases in the first months, but after 6 months its start to fall even here but is
still not significant statistically. The output tends to fall after a few months (Italy and
Germany there is an immediate, France after 4 months), with exception of Italy this fall is
statistically significant, and consistently with the transitory effects of a monetary contraction,
the output level shows evidence of a mean reversion to its pre-shock level. The exchange rate
significantly appreciates relative to the US dollar in all three countries following a monetary
contraction. In case of Germany the Federal Funds rate responds endogenously to the German
interest rate shock, probably because it's able to influence the global interest rate. Taking care
to this endogenous movement the DM/USD exchange rate response is stronger but
qualitatively similar. The impulse responses tend to be similar, no outstanding differences can
be recognized.
Kouparitsas [1999] uses the differences in the impulse response functions of the US regions
as a measure to show how different can reactions be to form a viable OCA6. His conclusion
that except two countries Ireland and Finland, EMU comes as close to being an OCA as the
U.S. does. His VAR for each country has four variables world oil price, aggregate EMU and
county income, and EMU region monetary policy (German short term interest rate), he uses
annual data and two lags, the identification is recursive with the ordering world oil price,
aggregate income, indicator monetary policy and county income, which means that the
monetary authority chooses the value of the monetary policy instrument after observing

 In this approach he follows Bayoumi and Eichengreen [1992], but this approach would only be a prove, if the
EMU countries would perform better, than the U.S. regions, because the U.S. is a functioning monetary union. If
EMU performs worse, it only means that it is not as much an OCA than the U.S. but it could turn out to be an
OCA after all.

contemporaneous movements in oil prices and aggregate output. This system makes it
possible to see how the different countries react to common shocks (monetary, oil price,
aggregate income) and idiosyncratic ones (country income). In case of common shock the
reaction should be similar, in case of idiosyncratic shock the reaction should be fast, as the
countries should be able to react flexible to them. Kouparitsas explains the different result for
the Ireland and Finland with their periphery stands.

Elbourne and De Haan [2004] searched for source, why different VAR estimates come to
different conclusions. They come to the conclusion, that most VARs are so different (different
set of countries, different variables, different identification method, different lag lengths etc.)
that it is no wonder that they come to different conclusion.

Guiso et al. [2000] criticize this literature to use different shocks (size and time path) and
different reaction function for the monetary policy in the different countries, which make it
impossible to make legitime comparisons among the responses. Second problem is the lack of
exchange rates. Third is the sensitivity of results on the identifying restriction as Kieler and
Saarenheimo [1998] showed.

3.2.2. Studies based on large-scale macroeconomic models
Using central banks' macroeconomic models to simulate the monetary transmission has the
advantage that its uses "insider wisdom", as the models are calibrated for the given economy.
On the other hand one can never be sure that the differences uncovered are because the
different models (according Guiso et al. [2000] different and arbitrary modelling choices) or
because the structural differences of the countries in the sample.

BIS (1995) used the central banks' macroeconomic models to see the effect of an common,
standardized monetary policy shock (1 percentage point increase, returning to baseline after
two years) in case a fixed floating and managed exchange rates. As stated before the results
are hard to compare, but finds that GDP responses are bigger in the bigger countries,
especially Italy, with a slightly larger and definitely longer lasting response. The price
response in Germany starts only after the second year but decreases ever after, and there is no
response in case of Austria.

3.3. Summing up.

In the next section I will summaries what changes were anticipated, and what changes had
happened after or because the EMU.

Guiso et al. [2000] draw the conclusion that the experiments that are most near to the ideal
model show the same result as the deductions of the indicator approach and this point to
noticeable differences in the transmission mechanism.

4. Effect of the common currency & monetary policy
After looking the differences found in transmission by other authors, let's turn to the changes
the common monetary policy could cause. According Arnold and de Vries [1999] the problem
with all the above cited literature is that, they concentrate on their "present" differences, and
disregard the fact that the monetary policy and the inflation in its control is in interaction with
the financial system, and through this the whole MTM will change because of the change in
the monetary regime. "Complete money market integration and the convergence of the
monetary transmission mechanism will come about as a by-product of the unitary inflation
regime." Arnold and de Vries [1999] p. 27. Searching for the changes the integrated monetary
policy and financial system could cause, Angeloni and Ehrmann [2003a] stresses that a
change in the monetary transmission mechanism should be a longer and slowly process,
which most probably began in the early 90's (with the Maastricht convergence), so it is not
sure, that there are breakpoints to find exactly at 1999. Suardi [2001] call attention to the fact
that on the one hand the strength of different channels can change due the regime change but
on the other hand at the same time the composition of different channels / importance of
different channels can alter as well, which makes the finding of changes even harder. Finding
changes aren't enough because the question Angeloni and Ehrmann [2003a] states "Has it
changed because of the EMU?" at page 471. has to be answered, so there are need for
counterfactual experiments with a control group to see if the changes found are a speciality of
the EMU or common changes around the world.

4.1. Indicators
Let's look through what changes the common monetary policy and currency could cause in
the connections pictured in figure 1., in this section I use indicators to show the different
fields of change.
First I look at the things that are going to change too slowly to be captures in this 10 years (or
17 years if the convergence is also included). According to Angeloni and Ehrmann [2003a]

heterogeneity caused by different consumer preferences (time preferences for example) or
differences in production7 should change slowly if they change anyway. Suardi [2001] adds
cross country differences of labour markets and firm size to the factors, in which changes will
not change rapidly. The next section is about things that probably changed because of the
regime change in the monetary policy.

4.1.1. Common monetary shocks from the common monetary policy (change in the aim,
instrument variable and in the aim  instrument segment)
The most striking change is the formation of the common monetary policy. In accordance
with the 105 article of the Maastricht Treaty, the aim of the European System of Central
Banks is to maintain price stability and "define and implement the monetary policy of the
Community". This means that the monetary policy got a single aim: price stability, defined by
the ECB as "year-on-year increase of the Harmonized Index of Consumer Prices for the euro
area, which does not exceed 2 percent in medium term". With harmonized instrumental
framework this result a single monetary shock for the whole euro area, which changes the
nature of the monetary policy signal, as it has new strategy and new euro area orientation. If
the volatility of monetary policy declines, the response of banks to a policy signal of any size
is likely to have increased Angeloni and Ehrmann [2003a].
To sum up the first step of the MTM was integrated institutionally.

4.1.2. Financial integration (change in the instrument variable  price of other financial
variables/assets segment)
"The market for a given financial instrument is considered fully integrated if all economic
agents with the same relevant characteristics acting in that market face a single set of rules,
have equal access and are treated equally." ECB [2005] p.5.
Highly integrated and deep financial markets enhance the efficiency of monetary policy in the
euro area by ensuring a smooth transmission of monetary impulses across all market
segments and countries. ECB [2007] p. 21.

The integration of the financial system could be the first thoroughly endogenous response in
the transmission mechanism. With the elimination of exchange rate risk the single currency

  See for example Krugman and Venables [1993] for the view that disappearance of the exchange rate risk will
lead to changes in the localization of production, to specialization, through which idiosyncratic shocks can come
to more relevance, and can according Dedola and Lippi [2005] "substitution theory" lead to a more differentiated
transmission mechanism.

removes one of the biggest sources of financial market segmentation (Angeloni and Ehrmann
[2003a]). As the segmentation disappears the increased competition (or even the threat of
contestibility) can ensure a convergence in the prices of financial assets and liabilities, as
inefficient banks are no longer able to pass extra cost on to costumers, so eventually would be
constraint to restructure or see their market share reduced8. So either through cross border
activity (foreign branches, mergers and acquisition (M&A) or foreign) or through
contestibility (and in even absence of cross border activity) the spread between the lending
and interbank rates and between the interbank and deposit rate should be reduced. This could
mean that even in case of an idiosyncratic shock, the agents of the economy would face the
same prices and dispose the same instruments to react which could help in a more flexible

Let's see what materialized from the above expectations. Since 2005 the ECB biannualy
publishes financial integration indicators9 mostly defined in Baele et al. [2004]. Using these
price (price dispersion), quantity (portfolio composition) and institutional indicators the ECB
report the state of integration in different segments of the financial intermediation. The
general diagnose is that "financial integration is more advanced in those market segments that
are closer to the single monetary policy, especially money market." ECB [2007] p.6.
Cross border interbank activity has been rising sharply, in absolute and relative terms and
relative to control cases (even non-area cases) but Angeloni and Ehrmann [2003b] couldn't
find any breakpoints in 1999. According ECB [2005] measures on the money market
(interbank short term debt and deposit) the cross country standard deviation dropped
significantly. ECB [2006] introduces new measures to be able to describe the infrastructural
surrounding that can further enhance the integration, in this market it is the number of large-
value payment systems, which decreased from 17 to 4, and it delivers bigger share of the inter
member state payments in volume and value. Angeloni and Ehrmann [2003b] states that the
banks holdings of cross border securities almost doubled, with an acceleration 1999, no
similar trend in the other EU countries or in the control group (this could influence the bank
lending channel). ECB [2006] shows that on the money market the cross-border holdings of
short-term debt securities issued by euro area residents further increased, by more in case

  On the other hand Angeloni and Ehrmann [2003a] states that the integration of the interbank market allows
relatively inefficient banks to access the better deposit raising technology of the foreign banks, helping them to

intra-euro area residents (from 7% (2001) to 11% (2006) of the portfolio) than extra-euro area
(from 2,8% (2001) to 3,7% (2006) in their portfolio).
On the government bond market the differences in the interest rates dropped, but mostly
during the convergence years (before 1999) and the yields tend to react more to common
factors (EMU wide news) instead of local ones (ECB [2005]). ECB [2006] finds the cross-
border holdings of long term debt securities issued by euro area residents increased from 11%
(1997) to 57% (2006) in the intra-euro area residents' international portfolios, versus the
increase from 4% (1997) to 9% (2006) in the extra-euro area residents' international
portfolios. In this case the infrastructure (security settlement systems (SSSs)) is still
Equity markets still reacts mostly on local factors, the euro area wide factors only explain
about 35% of the variance (the US or global factors about 15%), but it is rising and ECB
[2006] finds that the return dispersion between countries is declining (is even smaller than the
return dispersion between sectors), and so the diversification by country ought to be replaced
by sector based strategies.
Angeloni and Ehrmann [2003b] finds that cross border lending and deposit taking among euro
area countries, which was expected to increase after 1999, didn't changed dramatically. This is
line with findings of ECB [2005]: the relatively high dispersion of interest rates on consumer
credit, lending for house purchase on retail bank, which indicates that the retail banking sector
is highly fragmented, which is confirmed by the quantity based indicators because the cross
border retail bank lending activity in the euro area remains around 3.5 % of the total. The
infrastructure is segmented in case of retail payment systems as well, no sign of integration
can be found here, and maybe this segmentation is the source of the dispersion in the interest
rates ECB [2007]. Other source could be that in the retail banking sector the consolidation
inside the countries is more important than the cross-country consolidation (M&A the cross
border deals are below 30% to the total value of M&A, exception 2005 in which case it is
62%). There is almost no increase in share of assets of branches across the Euro Area (from
7,3% to 10,2%), counter the share of assets of the subsidiaries which increased from 3,76% in
2001 to 8,4% in 2007. Cross border branching shows no sign of increase in the euro area after
1999; it is very extensive in small countries and very limited in large ones, in line with the
control group (especially with the pre-1994 USA) Angeloni and Ehrmann [2003b]

Looking at the effect of integration on the interest rate pass-through Angeloni and Ehrmann
[2003a] investigate three aspects: the impact effect, the maximum effect and the time needed

to reach the maximum effect, in two time sample 1990-1998 and 1999-2002 (using a simple
regression and a VAR). They find a clear upward movement of both impact and the peak
response (except in Germany, which could be a sign that there was no policy signal
improvement in this case) and dispersion between countries declines. They use a control
group of UK, Sweden, Japan and USA, where they don't find such changes. Grouped by loan
and deposit type, they find that mortgage loans, long term business loans and long term time
deposits pass cross country homogeneity test, which implies that maturity matters. Angeloni
and Ehrmann [2003a] also measures the co-movements of the real interest rates (nominal
interest rate co-movement is assured by the arbitrage trades) with the variance of interest
differential, which shows a reduction mostly between 1990-94 and 1995-98 (the time of
convergence trades), so before the EMU and EU wide. According this measure the EU
reaches the integration level of the USA (control group).

4.1.3. Households
Angeloni and Ehrmann [2003a] maturities are growing with the inflation expectations shifting
to price stability (Spain and Italy maturity is on the rise, in France it is roughly constant, and
again in Germany the maturity is slowly declining (against the trend).
Arnold and de Vries [1999] also argue that the term that through the changes in the term
structure caused by the unification (very similar) of the inflation and the inflation
expectations/uncertainty, the whole capital market is on the way to be unified.

4.1.4. Non financial corporations
According to Angeloni and Ehrmann [2003a] the growing maturity is also characteristic for
non-financial corporations.

4.2 Quantitative
"The "ideal" study, based on past experience, which would be informative about differences
across countries in the transmission mechanism of a single monetary policy, would consider
the response of the various EU economies to the same temporal sequence of monetary policy
shocks, holding fixed the exchange rate among them." Guiso et al. [2000] p. 12.

4.2.1 Methodology
To solve the dynamics I estimate a 4 variable SVAR on three samples 1999:1 2007:12,
1992:1-2007:12 and 1980:1-2007:12 on monthly data.

The reduced form of the VAR model can be written in form:
                                             p             p
                                   xt  k   Bi xt i   Ci yt i  M  ut ,                     (1)
                                            i 1          i 1

where B1 ,..., B p are 4x4 matrices of coefficients, p is the number of lags, xt'  ipt cpit sit et 

is a vector of the four endogen variables namely: ipt the industrial production, rpit the price

level computed using the retail price index, sit the short term interest rate and ext the
exchange rate (in form domestic currency price of one unit of foreign currency) in that order.
C0 ,..., C p are 3x3 matrices of coefficients for yt'  iptf sit f cpt  , the vector of exogenous
                                                                       
variables, where iptf is the foreign industrial production, sit f the foreign short term interest

rate and cpt the commodity price index. M stands for the monthly dummies and u t is the
vector of reduced form residuals.
The main task after estimating equation (1) is to decompose residuals in u t into structural
shocks. This corresponds to finding the contemporaneous relationship between structural and
reduced form innovations, or finding the matrix A in the equation
                                                   ut  A1 t ,                                   (2)

where  t denotes the vector of structural shocks. The element in the i-th row and j-th column

of the matrix A1 is the magnitude by which the j-th structural shock affects i-th variable
simultaneously. As the matrix A1 is not unique, some additional information (in the form of
n  n 1 2 restrictions) and so the identification of structural shocks is needed. My
identification here follows that of Smets [1997] and Smets and Wouters [1999]10. In the main
part of the matrix A1 recursive identification is used (with the ordering as in vector x ), but it
is surely not appropriate between the shocks and residuals of the interest rates and exchange
rates. So in the beginning the matrix A1 takes the form:
                                               1    0           0      0 
                                                                          
                                                a    1           0      0 
                                        A1   21                           .                     (3)
                                               a31 a32          a33   a34 
                                                                          
                                               a41 a42          a43   a44 

To see what this mean, consider the following short-run reduced model as in Smets and
Wouters [1999]:
                                             utsi  a33 tP  a34 tE ,                            (4)
     See also Mojon and Peersman [2001].

                                                utex  a43 tP  a44 tE ,                    (5)

where  tP denotes the policy shock and  tE denotes the exchange rate shock (like a risk

premium shock), utsi and utex are the residuals of the equation of sit and ext in system (1).
Equation (4) can be considered as a short run reaction function of the monetary authorities.
Interest rate is adjusted to changes in the monetary policy stance  tP or to the movements of

the exchange rate due changes in risk premium or foreign interest rate shocks  tE . Equation
(5) states that in the equilibrium the exchange rate depends on domestic policy innovations
and exchange rate shocks.

The model in (4) and (5) is under-identified. The recursive scheme would imply that either
a34  0 or a43  0 . As the integrated capital markets are one of the main characteristics of the

globalization so to assume a43  0 would be too restrictive, actually because of the interest

rate parity the sign of a43 should be positive. The assumption of a34  0 would mean that the
monetary policy doesn't react to the shocks of the exchange rate within period, which seems
to be inappropriate for a small open economy. This problem of simultaneity should be
As my main interest is the identification of  tP , I solve the equation (4) and (5) for the
monetary policy shock:
                                              a44                     a34
                               tP                     utsi                   utex ,        (6)
                                       a44 a33  a43a34        a44 a33  a43a34

and following Smets [1997] renormalize the monetary policy shock such that the sum of the
weights on the domestic interest rate and exchange rate residuals gives one, so multiply both
            a44 a33  a43 a34
side with                     resulting:
               a44  a34

                                                   a44              a34
                                        tP              utsi            utex .             (7)
                                                a44  a34        a44  a34

Equation (7) can be rewritten as
                                             tP  1    utsi  utex ,                    (8)

where                 , (8) can be interpreted as a short-run monetary condition index (MCI)
               a44  a34

which is often used to measure the changes in the stance of monetary policy. In case of

depreciation the monetary policy, which aims to stabilize the exchange rate, would raise the
interest rate, so a34 should be positive. With a44  0 (because the exchange rate shock have a

direct positive effect on the exchange rate), the weight  should be between zero and minus
one. The change in this weight could be used as a measure of the monetary policy regime, as
it takes up the changes in the focus of the monetary policy. This weight can be calculated
from the estimated equation (with the assumption that  tP  0 )

                                       utsi   1    utex ,                                 (9)

using two stages least squares method with various instrumental variables to tackle the
endogeneity problems.

Equation (8) and the weight from equation (9) can be built in the identification through the
following equation
                                             AA'   ,                                         (10)
where   E  ut ut'  is the variance-covariance matrix of the reduced form and   E  t  t'  is

the variance covariance matrix of the structural shocks. It is assumed that the structural shocks
are orthogonal to each other, so the matrix  is diagonal. This is the estimated relationship
and the weight  is built into matrix A the following way
                                        1     0     0   0
                                                          
                                    A   21 1      0   0
                                                             .                                  (11)
                                         31  32 1    
                                                          
                                         41  42  43 1 

4.2.2 Results
"However it is important to bear in mind that EMU is a process, not a one time event. The
transition to a new currency and monetary policy was something economic agents had time to
prepare for, and adjust to, over a number of years. This complicates significantly the task of
identifying casual links." Angeloni and Ehrmann [2003b] p. 6.

What are the expected movements of the macro variable considered in this empirical model.
Monetary contraction should raise the interest rate and appreciate the exchange rate, these are
the general indications of tighter monetary policy stance. The initial rise in interest rate may

be reserved in the very short run due to deflationary pressure from monetary contraction, but
the initial impact should be an increase. Kim and Roubini [2000]
The price level should decline and the output should not increase. If the monetary policy is
identified correctly, so the monetary shock is really exogenous, and not a response on a future
inflation pressure (oil shock, foreign policy shocks), then no prices puzzle should appear. Kim
and Roubini [2000]
The reaction of the exchange rate to an interest rate increase depends on the expected
inflation, if there is a rise in the expected inflation then the interest rate movement is
accompanied by depreciation. The real question here if the increase is in the real or in the
nominal interest rate Kim and Roubini [2000]

According Elbourne and De Haan [2004] the more country the better it is, there shouldn't be
many variables and lags to save degree of freedom, but enough lags to make the residual
white noise. The longer the sample the better it is, but the chance of regime changes making
the coefficient estimates instable. The shock should be one standard deviation shock because
this shock, and its reaction should be in the data, but with an experiment with multiple
countries the shock should be the same in countries.

In the first experiment the sample is between 1999:1 and 2007:12, so the third stage of the
EMU. Data for the different countries and different variables are in the A appendix. After
estimating the equation system in (1) with p  5 (5 lags are enough to take care of the
autocorrelation) and estimating the equation (9) (   0, 022367 ) a 0,0025 monetary policy
shock was introduced. The reaction of the different variables can be found in figure 3. In case
of Germany, France and Italy additional shocks were introduced that the path of the interest
rate and the exchange rate is same as in case of the EMU experiment.

Figure 3; Impulse response functions, 0,0025 EMU monetary policy shock, with constraint
exchange rate and interest rate paths.

      0.01                                                0.002
     0.008                                               0.0015
     0.006                                                0.001
                                              EU-ip                                                 EU-cpi
                                              D-ip       0.0005                                     D-cpi
                                              F-ip              0                                   F-cpi
    -0.002 1   5   9 13 17 21 25 29 33        I-ip       -0.0005 1       5    9 13 17 21 25 29 33   I-cpi
    -0.004                                                -0.001
    -0.006                                               -0.0015

    0.003                                                0.02
    0.002                                                0.01
                                               EU-si                                                 EU-ex
    0.001                                                   0
                                               EU-si                                                 EU-ex
        0                                               -0.01 1      5       9 13 17 21 25 29 33
                                               EU-si                                                 EU-ex
    -0.001 1   5   9 13 17 21 25 29 33                  -0.02
                                               EU-si                                                 EU-ex
    -0.002                                              -0.03
    -0.003                                              -0.04


5. Conclusion

6. Literature
ANGELONI, IGANZIO – EHRMNANN, MICHEAL [2003a]: Monetary Transmission Evidence. Monetary transmission
      in the euro area: early evidence. Economic Policy, Vol. 18., No. 37., (October, 2003) pp. 469-501.

ANGELONI, IGANZIO – EHRMNANN, MICHEAL [2003b]: Monetary policy in the Euro Area: any changes after
      EMU? European Central Bank, Working Paper Series, WP No. 240., (July, 2003).

ARNOLD, IVO J.M. – DE VRIES, CASPER G. [1999]: Endogenous Financial Structure and the Transmission of ECB
      Policy. Tinberger Institute, Discussion Papers No. 99-021/2 (March 1999).

        Measuring Financial Integration in the Euro Area. European Central Bank, Occasional Paper Series, OP
        No. 14., (April 2004)

BAYOUMI, TAMIM – EICHENGREEN, BARRY [1992]: Shocking Aspects of European Monetary Unification.
      National Bureau of Economic Research, NBER Working Paper Series, WP No. 3949., (January, 1992).

BERNANKE, BEN S. [2007]: Financial Accelerator and the Credit Channel. speech delivered at The Credit
      Channel of Monetary Policy in the Twenty-first Century Conference, Federal Reserve Bank of Atlanta,
      Atlanta, June 15, 2007. download:

BERNANKE, BEN – GERTLER, MARK [1995]: Inside the Black Box: The Credit Channel of Monetary Policy
      Transmission. The Journal of Economic Perspective, Vol. 9, No. 4, (Autumn, 1995), pp. 27-48.

CECCHETTI, STEPHEN G. [1995]: Distinguish Theories of the Monetary Transmission Mechanism. Federal
       Reserve Bank of St. Louis, Proceedings (May, 1995), pp. 83-97.

CECCHETTI, STEPHEN G. [1999]. Legal Structure, Financial Structure, and the Monetary Policy Transmission
       Mechanism. Federal Reserve Bank of New York, FRBNY Economic Policy Review, (July, 1999), pp.

COCHRANE, JOHN H. [1998]: What do the VARs mean? Measuring the output effects of monetary policy. Journal
      of Monetary Economics, Vol. 41., (1998), pp. 277-300.

DEDOLA, LUCA – LIPPI, FRANCESCO [2005]: The monetary transmission mechanism: evidence from the
      industries of five OECD countries. European Economic Review, Vol. 49., No. 6., (August, 2005), pp.

ECB [2005]: Indicators of Financial Integration in the Euro Area. European Central Bank, Frankfurt,
      (September, 2005).

ECB [2006]: Indicators of Financial Integration in the Euro Area. European Central Bank, Frankfurt,
      (September, 2006).

ECB [2007]: Financial Integration in Europe. European Central Bank, Frankfurt, (March, 2007).

EHRMANN, MICHAEL [1998]: Will EMU Generate Asymmetry? Comparing Monetary Policy Transmission
      Across European Countries. European University Institute, Department of Economics, EUI Working
      Paper ECO No. 98/28., (Sept, 1998).

ELBOURNE, ADAM – DE HAAN, JAKOB [2004]: Asymmetric Monetary Transmission in EMU: The Robustness of
      VAR Conclusion and Cecchetti's Legal Family Theory. CESifo Working Paper Series, Category 6.
      Monetary Policy and International Finance, WP No. 1327., (November, 2004)

FRANKEL, JEFFREY A. – ROSE, ANDREW K. [1996]: The Endogenity of the Optimum Currency Area Criteria.
      National Bureau of Economic Research, Working Paper Series, WP No. 5700., (Aug., 1996).

GUISO, L. – KASHYAP, A. K. – PANETTE, F. – TERLIZZESE, D. [2000]: Will a Common Eurpean Monetary Policy
        Have Asymmetric Effects? Banca d'Italia, Temi di discussione del Servizio Studi, No. 384., (October,

HM TREASURY [2003]: EMU and the monetary transmission mechanism. downloaded:
     hire.cfm 15th March 2008. 12.12

KASHYAP, ANILL K. – STEIN, JEREMY C. [1997]: The role of banks in monetary policy: A survey with
      implications for the European monetary union. Federal Reserve Bank of Chicago, Economic
      Perspective, (Sept, 1997), pp. 2-18.

KIELER, MADS – SAARENHEIMO, TUOMAS [1998]: Differences in Monetary Policy Transmission? A Case not
        Closed. European Comission, Directorate-General for Economic and Financial Affairs, European
        Economy – Economic Paper Series No. 132., (November, 1998).

KIM, SOYOUNG [1999]: Do monetary policy shock matter in the G-7 countries? Using common identifying
       assumptions about monetary policy across countries. Journal of International Economics, Vol. 48., No.
       2., (August 1999), pp. 387-412.

KIM, SOYOUNG – ROUBINI, NOURIEL [2000]: Exchange rate anomalies in the industrial countires: A solution with
       a structural VAR approach. Journal of Monetary Economics, Vol. 45., No. 3., (June 2000), pp. 561-586.

KOUPARITSAS, MICHAEL A. [1999]: Is the EMU a viable common currency area? A VAR analysis of regional
      business cycles. Federal Reserve Bank of Chicago, Economic Perspective, Vol. 23., No. 4., pp. 2-20.

KRUGMAN, PAUL – VENABLES, ANTHONY [1993]: Integration, Specialisation and Adjustment. National Bureau
      of Economic Research, Working Paper Series, WP No. 4559., (Dec., 1993).

MISHKIN, FREDERIC S. [1996]: The channels of monetary transmission: lessons for monetary policy. National
       Bureau of Economic Research, NBER Working Paper Series, WP No. 5464., (February, 1996).

MOJON, BENOIT [2000]: Financial structure and the interest rate channel of ECB monetary policy. European
       Central Bank, Working Paper Series, WP No. 40., (November 2000).

MOJON, BENOIT – PEERSMAN, GERT [2001]: A VAR description of the effects of monetary policy int he
       individual countries of the euro area. European Central Bank, Working Paper Series, WP. No. 92.,
       (December, 2001).

MUNDELL, ROBERT A. [1961]: The Theory of Optimum Currency Area. The American Economic Review, Vol.
      51., No. 4., (Sept., 1961), pp. 657-665.

PEERSMAN, GERT [2001]: The Transmission of Monetary Policy in the Euro Area: Implications for the European
       Central Bank. Universitait Gent, 2001. March

RUDEBUSCH, GLENN D. [1998]: Do Measures of Monetary Policy in a Var Make Sense? International Economic
      Review, Vol. 39., No. 4., Symposium on Forecasting and Empirical Methods in Macroeconomics and
      Finance. (Nov., 1998), pp. 907-931.

SIMS, CHRISTOPHER A. [1992]: Interpreting the macroeconomic time series facts. The effects of monetary policy.
        Cowles Foundation Paper, No. 823., North-Hollnad, European Economic Review, Vol. 36., No. 5.
        (June, 1992), pp. 957-1011.

SIMS, CHRISTOPHER A. [1998]: Comment on Glenn Rudebusch's "Do Measures of Monetary Policy in a Var
        Make Sense?". International Economic Review, Vol. 39. No. 4., Symposium on Forecasting and
        Empirical Methods in Macroeconomics and Finance, (Nov., 1998), pp. 933-941.

SMETS, FRANK [1997]: Measuring Monetary Policy Shocks in France, Germany and Italy: the Role of the
       Exchange Rate. Bank for International Settlements, Monetary and Economic Department, Working
       Paper Series, WP No. 42. (June, 1997).

SMETS, FRANK – WOUTERS, RAF [1999]: The Exchange Rate and the Monetary Transmission in Germany. De
        Nederlandsche Bank, DNB Staff Reports, No. 35., (1999).

SUARDI, MASSINO [2001]: EMU and asymmetries in monetary policy transmission. European Comission,
       Directorate-General for Economic and Financial Affairs, Economic Paper No. 157., (July 2001).

TAYLOR, JOHN B. [1995]: The Monetary Transmission Mechanism: An Empirical Framework. The Journal of
       Economic Perspective, Vol. 9, No. 4, (Autumn, 1995), pp. 11-26.

TOBIN, JAMES [1969]: A general equilibrium approach to monetary theory. Journal of Money, Credit and
        Banking, Vol. 1., No. 1., (February 1979), pp. 15-29.

A appendix
                 variable   country   time series
                                      Industry - monthly data, Production index - Data adjusted by working
                                      days (WDA), Euro area (EA11-2000, EA12-2006, EA13-2007, EA15),
                            EMU       Index, 2000=100, Not seasonally adjusted data, Total industry
                                      (excluding construction)
                                      source: EUROSTAT
                                      UUNI98 Production / Mining, quarrying and manufacturing /
                                      seasonally adjusted 2000=100 General: 1 Source of the original data:
                 ipt                  Federal Statistical Office. - 2 Territorial definition: Germany as of
                                      1991, prior to that western Germany (chain-linked over annual average
                                      of 1991
                                      source: Bundesbank
                                      13266..CZF... INDUST PRODUCTION, SEAS ADJ index
                                      source: IMF IFS database
                                      13666..CZF... INDUST.PRODUCTION,SEAS.ADJ. index
1999:1-2007:12                        source: IMF IFS database
EMU                                   Harmonized consumer price index, 2005=100, Euro area (EA11-2000,
                                      EA12-2006, EA13-2007, EA15), Not seasonally adjusted data, HICP -
                                      All items (HICP=Harmonized Index of Consumer Prices)
                 cpit                 source: EUROSTAT
                                      Harmonized consumer price index, 2005=100, Not seasonally adjusted
                            D, F, I   data, HICP - All items (HICP=Harmonized Index of Consumer Prices)
                                       source: EUROSTAT
                                      Interest rates - monthly data (mf-3mi-rt), Not seasonally adjusted data,
                            EMU,      3-month interest rates (average), Euro area (EA11-2000, EA12-2006,
                            D, F, I   EA13-2007, EA15)
                                      source: EUROSTAT
                                      ECU(Euro)/ United States Dollar Euro exchange rates - Monthly data
                            EMU,      Average National currency (including 'euro fixed' series for euro area
                            D, F, I   countries)
                                      source EUROSTAT
                                      US industrial production index (INDPRO; Industrial Production Index
                            EMU,      G.17 Industrial Production and Capacity Utilization), seasonally
                            D         adjusted, 2002=100
                                      source: Federal Reserve Bank of St. Louis
                                      German UUNI98 Production / Mining, quarrying and manufacturing /
                                      seasonally adjusted 2000=100 General: 1 Source of the original data:
                                      Federal Statistical Office. - 2 Territorial definition: Germany as of
                            F, I
                                      1991, prior to that western Germany (chain-linked over annual average
all                                   of 1991
                                      source: Bundesbank
                            EMU,      US 3-month monthly money market interest (irt_st_m; mat_m03m)
                       f    D         source: EUROSTAT
                                      German 3 month money market interest rate, monthly average
                            F, I
                                      source: Bundesbank
                                      commodity non-fuel price index (PNFUELW), which includes food,
                 cpt        all       beverages and industrial inputs price indices, 2005=100
                                      source: IMF


Shared By: