Jurnal Internasional Koperasi
Monetary
Condition Index with Time Varying Weights: An Application to Turkish Data
İlyas Şıklar
Anadolu
University, Faculty of Economics and Administrative Sciences
Department of
Economics
E-mail:
isiklar@anadolu.edu.tr
Burhan Doğan
Anadolu
University, Faculty of Economics and Administrative Sciences
Department of
Economics
E-mail:
burhand@anadolu.edu.tr
Received:
January 16, 2015 Accepted: March 7,
2015
doi:10.5296/ber.v5i1.7101 URL:
http://dx.doi.org/10.5296/ber.v5i1.7101
In the study carried out by Hyder and Khan (2006) weights
for construction of MCI are obtained by using Johansen’s cointegration t
loosing periods and 8 tightening periods in terms of monetary conditions in
Pakistan between
Abstract
In this study, the evaluation of the
impact of the fluctuations in interest rate and exchange rates on monetary
policy is carried out through the use of Monetary Conditions Index (MCI). The
weights for construction of MCI are derived using the time varying framework
with Kalman Filter algorithm. Despite the simplifying assumptions made during
its construction, the time varying version of MCI provides opportunities for
the analysis of the contribution of the monetary conditions to the evaluation
of Turkish economy, especially in recent past when
it is compared to traditional constant
weighted MCIs. According to the fundamental result obtained in the study, the changes in inflation give rise
to changes in both interest rates and exchange rates. Moreover, the result
obtained also demonstrates that the interest rate channel, when compared to
exchange rate channel, have a stronger and more rapid impact on the transfer of
the changes in policies to economy.
Keywords: Monetary
policy, Interest rate, Exchange rate, Monetary conditions index, Kalman filter
algorithm.
1.
Introduction
Monetary policy generally affects
economic conditions such as inflation and aggregate output through many ways
known as the transmission mechanism. Sudden changes in political tools cause
changes in investment, saving and spending habits of consumers and investors
through interest rate. Likewise, differentiation of the value of the exchange
rate together with sudden change in monetary policy cause change in import and
export data. Occurrence of such facts in tandem may cause some effects on
aggregate demand, aggregate output and prices by the way of consumption,
investment and net export. Because the effect of interest rate and exchange
rate channel on economic conditions is high, an index providing weighted mean
has been generated. Such an index known as monetary condition index (MCI) can
be defined as weighted mean of changes in exchange rate and interest rate
dependently values of them in a basic period (Batini and Turnbull, 2002).
Because of the importance of both variables in monetary transmission, MCI shows
changes in short run interest rate and exchange rate as a single variable. Thus
influence degrees of interest rate and exchange rate on aggregate demand and
general price level can be determined by calculating the MCI.
MCIs are motivated in terms of
standard Keynesian open economy characterized by aggregate demand among which
real interest rate and real exchange rate are as a function in others
(Bayangos, 2000, pp 6). Because influences of interest rate and exchange rate
on the economy change according to magnitude and openness of the economy, MCI
measurements calculated in various countries may differ. In small open
economies, monetary policy may be quite effective on general level of aggregate
output and prices through interest and exchange rate. Increasing the cost of
financial capital out of productive factors by high interest rates may cause
decreases at the level of aggregate output. This case may generally cause
appreciation of local currency while other conditions are constant. The
appreciated local currency will cause decrease in net export and decrease in
domestic production (Korhonen, 2002, pp 4). Because of this interaction
monetary authorities in the small open economies may frequently utilize
interest and exchange rate tools in order to be effective on output and prices.
By finding weighted mean of interest rate and exchange rate within the MCI,
interventions of monetary authorities to aggregate demand and prices may be
made more effective. Therefore, a central bank can optimize its objective
function through the regulation of weighted average comprised of exchange rate
and interest rate in accordance with the existing macroeconomic conditions in
the country (Gerlach and Smets, 2000).
MCI can be used as the monetary indicator of inflation targeting monetary
policy. Such an indicator shows whether central banks loosen or tighten
monetary policy. For this reason MCI as a monetary policy indicator has an
important role as operational target in conducting the monetary policy
(Freedman, 1995). Occurrence of an increase in MCI indicator signals that there
is a tightening in monetary policy and a decrease in MCI indicator signals a
loosening in monetary policy. Since most of the studies in the literature have
obtained MCI as 2:1, a change in exchange rate by 2 % and a bustle in interest
rates by 1 % causes an equivalent effect on aggregate demand and prices (Brash,
1997). Hence we can generate following table:
In
this study, MCI calculation is carried out as to include the years of 1991
–2013 in monthly basis. In calculation of MCI, almost all studies concerning
Turkey have been carried out by using constant weights for interest rate and
exchange rate variables as common in the literature. However, this study
differs by using the time varying coefficients for the mentioned variables in
construction process of MCI for the period under investigation. Through the
calculation of MCI with the time varying coefficients in Turkey, opportunity
for being able to see influence levels of monetary authorities, institutions
and corporates on changes in monetary policies will be more easily provided. In
the first part, MCI studies carried out all around the world are included and
differences between countries are presented. In the second part, information
about limits and reasons of the model to be applied will be given. In the third
part, results are interpreted by carrying out empirical analysis. In the last
part, study is going to be summarized and evaluated in general terms and
empirical results.
2. Literature
Today,
as an indicator of monetary policy conditions, MCI is measured by central banks
and various institutions of many countries (Neil R. Ericcson et al., 1998, pp
1). Today, calculated MCI values that are used as an indicator of monetary
policy under the leadership of Central Bank of Canada are published by central
banks, public institutions and companies (Gerlach and Smets, 2000, pp 1680).
Whereas MCI is calculated by institutions such as OECD, IMF in addition to
central banks in order to evaluate monetary policy, companies such as Deutsche
Bank, Goldman Sachs, JP Morgan and Merrill Lynch have published MCI for the
purpose of investigating monetary conditions in many countries (Ericsson et
al., 1998, pp 2).
When
we look at previous studies concerning with MCI, some of the remarkable studies
conducted are Duguay (1994) for Canada, De-Simone, Dennis, and Redwards (1996)
for New Zealand, Jore (1994) for Norway. Since 1990, Canadian Central Bank has
been using MCI updated monthly as operational target instead of monetary
aggregates. Structure of MCI generated by Canadian Central Bank was based on
quarterly output estimations of Duguay (1994) covering the period of 1980-1990.
Short-term real and nominal MCI was estimated through independent variables
including real exchange rate, real interest rate and real output. In the
mentioned study, whereas changes in interest rates affect aggregate demand by
3%, changes in exchange rates affect 1%. MCI for New Zealand was based on a
model generated by De-Simone, Dennis, and Redwards (1996) including 1987-1995
period. New Zealand Central
Bank has used real
interest rate and real exchange rate as independent variables and real output
as dependent variable. MCI generated for Norwegian Central Bank was based on
aggregate demand model with single-equation by using quarterly data for the
period 1985-1994. When we look at the results obtained in this study, MCI weights
are 2:1(Eika et al., 1996).
Frochen
(1996) has generated monetary condition index for five European countries
(France, Germany, Spain, England, Italy) with the series including nominal
short and long term interest rates and effective exchange rate. Indicators
showed that monetary policy may have a balancing effect on price level in
France and Germany since 1990. Hataiseree (1998), in his study regarding
Thailand, indicated that usage of MCI index presents a positive image as an
important indicator for characterizing short term monetary conditions in
conducting monetary policy as well as evaluating the behavior of inflation
rate. However, he draw attention that further investigations are required for
being able to guide monetary policy under floating exchange rate regime that
would occur in future. The empirical analysis carried out in this study
demonstrates the relative significance of exchange rate and interest rate
variables in the construction of MCI. It is particularly underlined that the
use of MCI in the evaluation of economic and monetary conditions in Thailand is
more effective compared to the use of exchange rate or interest rate
individually. The study includes the empirical analysis of the relationship
between inflation rate and MCI for Thailand economy. Empirical results show
that there is a high correlation between general evolvement of MCI and
inflation. Another important reflection of this finding for conducting monetary
policy is that emergence of MCI is a useful indicator for monetary conditions
in the short term and can be used for affecting oncoming evolvements of
inflation (Hataiseree, 1998, pp 13).
Kesriyeli
and Koçaker (1999) have generated weights through estimation of price equation
rather than aggregate demand equation due to being impetus of exchange rate in
price regulation process in Turkey. Moreover, wei between operational target
and final target. Gerlach and Smets (2000) have generated a theoretical
model showing that monetary condition index may have been written in the aspect
of suitable feedback rule of central bank. They estimated reaction functions of
central banks to changes in exchange rate by targeting inflation for Australia,
Canada and New Zealand. Whereas the central banks of New Zealand and Canada
using the MCI as operational target react to changes in exchange rate potently,
Australian central bank does not.
Korhonen
(2002) estimated monetary transmission mechanism in three European Union
candidate countries (Czech Republic, Poland and Slovakia). MCI including
exchange rate which has very little effect on Slovak economy bewilderingly was
calculated. While MCI generated for Czech Republic in the study produces quite
similar results with small EU countries, Poland seems extremely sensitive to
exchange rate changes.
Siklos
(2002) mentions advantages and disadvantages of MCIs. He indicated that while
generated MCI may increase credibility of monetary policy, a central bank’s cl
each change in MCI may cause confusion among financial market participants.
Advantages and disadvantages of MCI are also mentioned by KaytancıShe
points(2008)outthatthe.usage of a model including unnecessary externality
assumptions and consisting of non-stationary parameters in the assessment of
monetary policy may be misleading.
In the study carried out by Hyder and Khan (2006) weights
for construction of MCI are obtained by using Johansen’s cointegration t
loosing periods and 8 tightening periods in terms of monetary conditions in
Pakistan between
March
1991 and April 2006. Knedlik (2006) utilized MCI to evaluate the optimality of
the monetary policy in South Africa. Gan (2008) estimated comparative
effects of exchange rate and interest rate on output gap by using weights of
real exchange rate and real interest rate in the estimation of optimal monetary
policy.
3.
Generating MCI
In
order to be able to generate monetary condition index various econometric
techniques can be used. As observed in previous studies, measurement of MCI has
been handled in two aspects. According to this, MCI can be established by
taking into account the impact of changes in interest rate and exchange rate on
the "aggregate demand" or "price level". In the first case,
the weights used for the construction of MCI are obtained through the
prediction of aggregate demand equation. Whereas in the second case, the impact
of changes in exchange rate and interest rate on prices is analyzed. In this
case, the weight of exchange rate on MCI is bigger because of its direct impact
on the prices in addition to its indirect impact resulting from the aggregate
demand (Kesriyeli and Koçaker, 1999, pp 3). It should be noted that such
measurements can be carried out through nominal and real values (Freedman, 1995).
3.1
Methods
When
we look at first studies carried out about MCI in the literature, models based
on methods with single equation are used, however in the studies carried out
recent years empirical analysis are carried out through using multi equation
models. Furthermore, by using structural VAR models, MCI can be generated
without any necessity for restrictions used in other simultaneous equations
(Tarı and Bozkurt,-5).Asthevalue2006,ofMCIat timeppt is 4defined as the
weighted sum of the changes in exchange rate and interest rate in the selected
base year, the formula to be used in the calculation of MCI can be written as
follows:
In
this equation, IRt
and ERt
refer respectively to interest rate and exchange rate at time t. IRb
and ERb
are the interest rate and exchange rate in the selected base year, respectively.
The most important factor in the formation of this equation is the value of
weights (ws),
since the value of the weights in question demonstrates the relative
significance of exchange rate and interest rate in respect of impacting the
final objective either in the form of inflation or output level. The weighing
of the two variables in the construction of MCI is possible through the use of
various econometric techniques. Among these techniques, the most frequently
used are: (i) approach based on the use of a single equation in relation
with either the price level or output level, (ii) approach based on
trade elasticities, (iii) approach based on Vector Autoregressive (VAR)
and Johansen cointegration models. As discussed in the final section, the use
of VAR and Johansen cointegration models is more preferable considering the
shortcomings associated with the first
two approaches.
The shortcomings mentioned include omitted variable bias, dynamic exogenity and
feedback problems. The cointegration methodology is the generally preferred
approach as it takes the aforementioned problems in consideration.
3.2
Restrictions
Although MCI can serve as an
important indicator of monetary conditions, usage of MCI as an operational tool
is restricted because selection of weights for interest rate and exchange rate
depends on its relative effect on inflation and aggregate demand. Other
restrictive factors are that MCI does not prevail on market expectations on
policies and selection of variables of MCI may change. Such kind of reasons
restricts usage of MCI as a monetary condition indicator (Hyder and Khan,
2006). Usage of MCI as an operating target in monetary policy is found
inconvenient due to extreme changes in its components. In some countries
(Canada, New Zealand etc.), MCI is used as operating target in conducting
monetary policy whereas in some countries (Norway and Sweden) there are
restrictions on its usage as a target in monetary policy design (Neil R.
Ericsson et al., 1998).
If interest rates and buying rate of
exchange are completely controlled by monetary authority, the importance of
monetary conditions would become relatively simple. In practice, as Kennedy and
Van Riet (1995) state, differentiation of these variables highly depends on
reaction of the market for expected monetary policy stance.
Since the beginning of 1990, whereas
central banks have estimated monetary condition index as either an operating
target of monetary policy or an indicator of monetary conditions, it is
observed that problems with MCI estimation have been increasing. For that
reason, when we want to evaluate effects of monetary policy on the economy, we
should evaluate the links between economic fundamentals and external factors
like exchange rate (Stevens, 1998). When we consider such kind of factors, it
is suggested that MCI should not be used alone and other channels and effects
of lags should not be neglected in designing monetary policy toward specified
targets like inflation (Svensson, 1998).
4.
Constructing the MCI For Turkey
As it was previously pointed out by
the authors of the cited works, the construction of monetary conditions index
is not an easy task, since such a variable not only needs to reflect the
current developments in financial markets but also to constitute a meaningful
indicator in relation with the future economic activities. Furthermore, a
correctly predicted monetary conditions index is also expected to "provide
a continuously updated flow of information on the future whereas the
traditional economic predictions are updated on a monthly or quarterly
basis" (Mayes and Viren, 2001, pp 8).
Generally speaking, MCI is an
indicator which provides information on the current inflation and monetary
policy in a given country. Nevertheless, as Grande (1997) points out, how the
required information will be obtained from a composite indicator does not
constitute the only problem. The additional assumptions required in order to
put this indicator in practice constitute another problem. In this section, we
are going to construct an indicator with above-mentioned characteristics.
The first stage in the analysis is the creation of an
aggregate measure for MCI. Following Goodhard and Hofmann (2001), we will
develop our analysis within the context of two assets model, namely short-term
interest rate and real effective exchange rate. In this part, we are going to
explain how MCI is derived and especially how it is used in the formulation of
monetary policy. The first problem in relation with the construction of MCI is
the determination of weights in relation with each asset. Goodhard and Hoffman
(2001) suggest three different approaches in this regard: the estimation of a
large-scale macro econometric model, the estimation of an equation system
consisting of reduced form aggregate demand and aggregate supply functions and
the analysis of impulse response functions obtained through VAR model
estimation. The authors mentioned make a reference to the difficulties posed by
the first model and conclude that the preference of the second and third
methods would be more appropriate. The two methods offer similar results except
for Germany and England. However, a common problem for the two proposed methods
attracts attention: despite the size of the sample used, the weights used in
relation with each financial variable are fixed. As a matter of fact, the
portfolios of firms and households tend to change in relation with business
cycle and the emergence of certain events in the economy. In this study, we
will try to overcome this problem by proposing an alternative for the
calculation of weights in relation with each variable. In order to consider the
changes manifested by the weights over time, what we propose is the use of
Kalman filter algorithm.
Taking into consideration the
contributions made by Eika et al.(1997), Mayes and Viren (1998), Goodhard
(2000), Goodhard and Hofmann (2001) and Mayes and Viren (2001) we are going to
formulate a formal model which demonstrates the importance of financial
variables in the implementation of monetary policy. By this means, we can form
a simple model corresponding to the traditional backward looking aggregate
demand - aggregate supply model which has been enhanced to include assets
market (Goodhard and Hofmann, 2001):
Where
πtrefers
to inflation and is calculated as 100×[ln(CPIt/CPIt-12)]
while yt,
reflecting the output gap, is the gap between the actual and potential output and
is calculated as percentage deviation of natural logarithm of monthly
industrial production index from its Hodrick-Prescott trend. The financial
markets are represented by two variables: ri and re. These
variables reflect respectively the gap between the real and potential interest
rate and the real effective exchange rate. Long term values of asset prices
have been calculated through the use of Hodrick-Prescott filtering technique.
For the selection of sample, the necessity to cover the principal events
causing changes in government and monetary policies was taken as basis. In this
case, the period between January 1991 and December 2013, during which
fundamental transformations and changes occurred in terms of politics and
economic policy, was selected and 276 observations on a monthly basis were
included.
The
construction of MCI was divided into two phases: the first phase covers the
estimation of Equation 2 through the use of Kalman filter algorithm, whereas
the second phase includes the definition of composite index to be formed by
using coefficients which change depending on time. For the purposes of the
analysis in this study, the most important point is the values of the coefficients
in Equation 2. Therefore, in order to understand the potential significance of
an unobservable change in βi,t, the
estimation of Equation 2 in state space form is required. Consequently,
Equation 2 can be rewritten in state space form as follows:
In
these equations, Z reflects a (T x k) dimension matrix containing
all the explanatory variables and a constant term, whereas yt
reflects the value of output gap, as explained above. On the other hand, βtis
(kx1) dimension state vector covering all the slope coefficients which
change over time. The (kxk) dimension F matrix contains the
autoregressive coefficients of βt.
By doing so, we allow the coefficient βtto
follow random walk process. It is assumed that the error
terms
are independent white noise and Var(µ) = Q;
t
By
stating Equation 2 as above, the estimation of state vector for t = k+1;
k+2; ::: T through the use of Kalman filter is made possible. Considering
the purposes of our analysis, this algorithm enables us to observe the dynamics
of the iteration between the economic activity gap and the variables which
explain this gap. This technique is effective even in cases where we suspect a
structural break in the estimation period but are not sure of its exact date.
In fact, this recursive technique is the calculation of the linear least
squares of the estimated state vector when the observation at time t is
as given. Whereas the initial value of state coefficient estimated by OLS is
given, the state coefficient of each term is updated on the basis of inflation
in the previous terms by making use of this technique. Therefore, likelihood
function is maximized until reaching convergence value. The following equations
are recursively
Where Ht
is defined as FPt-1/k-1F’Rand
βt+1/t
is the estimation of state vector for t+1 period based on the
information at time t.
The
methodology summarized above provides an opportunity for removing the problems
resulting from an unobservable factor affecting the output gap. Therefore, it
is possible to observe the change of asset price weight over time for each
endogenous variable in the model.
The second phase consists of the
calculation of the weight of each asset within MCI. The equation to be used to
this end has been given below:
Here, Xi,t
is the price of asset i at time t, whereas n reflects the
size of the sample. Weights which change over time as calculated through the
use of above equation have been set out in Figure 1.
Weights
for real interest rate Weights
for real exchange rate
Figure 1.
Weights Estimated for Calculation of the MCI
The last
phase in the calculations is the definition of MCI:
According to this equation, the
increase in MCI means the tightening of monetary conditions. The Figure 2 below
shows the MCI values for Turkey calculated through time varying weights as
basis. The calculated MCI fluctuates between dramatically different ranges.
Taken as a whole, MCI can be said to fluctuate around (-1) throughout the
period under consideration. Moreover, the volatility of MCI is observed to have
increased especially between the years 2000 and 2003 as well as 2007 and 2009.
Turkey
faced a chronic and increasing inflation problem beginning from the 1970's up
until the end of 1990's. During this period, the monetary policy was changed
frequently and after 1990, the monetary policy implemented in Turkey targeted
the real exchange rate. In this period, nominal exchange rate lost value in
accordance with inflation to maintain real exchange rate fixed and ensure the
competitiveness of Turkish exporters. Nevertheless, the policies implemented
are not transparent. Frequent changes in policies and lack of transparency in
exchange rate policies brought about a significant loss of credibility for
Central Bank of the Republic of Turkey (CBRT). For instance due to huge budget
and current account deficits in 1994, domestic currency had been devaluated
almost 60 percent against US Dollar. The shaded area in Figure 2 depicts the
tightening monetary conditions before 1994 twin deficits crisis in Turkey. On
the other hand, Turkey faced a rapid financial outflow in November 2000 as
foreign investors’lossoftrust.After the obstacles before the capital flow were removed
in 1990, the underdeveloped financial system was left vulnerable against
speculative attacks. The financial outflows mentioned resulted in a domestic
banking crisis and emergence of a vast budget deficit, which in turn paved the
way for a rapid monetary expansion, increasing inflation and rapidly
depreciating national currency. The second shaded area in Figure 2 shows the
unbearable tightening of monetary conditions before February 2001. As the
exchange rate target became unsustainable, the domestic currency was left for
fluctuation. In the post-crisis era, reforms aiming at restructuring Turkish
economy were enforced. In addition to transition to floating exchange rate
regime, institutional reforms aiming at increasing the role of the markets,
restructuring public banks and decreasing the influence of public sector on the
economy were implemented. One of the most important reforms enforced during
this period is the legal amendment which has enabled the Central Bank to
enforce an independent monetary policy. Thanks to this reform, prerequisites
for transition to a monetary policy aimed at inflation targeting were provided.
The process of transition to inflation targeting began in 2002 and ended in
2006. After this, the official monetary policy of Turkey was transformed to
inflation targeting regime. Another
objective of the structural reforms implemented was to accelerate the
harmonization of Turkish economy with European Union standards.
The
transition to floating exchange rate regime in Turkey decreased the impact of
exchange rate pass-through to prices to a great extent and weakened the issue
of dollarization. It is possible to assert that this change in exchange rate
policy was reflected positively in the effectiveness of monetary policy. On the
other hand, considering that Turkey is a small open economy, the transition to
floating exchange rate regime caused a change in monetary transmission
mechanism by strengthening interest rate and credit channels as well as
weakening the effect of exchange rate pass-through. This transformation is also
supported by the economic theory. According to the theory, flexible foreign
exchange rate regime provides public administration with opportunities in
taking the necessary measures for tackling domestic problems such as domestic
shocks and to implement a flexible monetary policy (Calvo and Mishkin, 2003).
As
a part of 2001 reforms, CBRT obtained instrument independence and thus was
freed from the impact of political decisions and choices. In addition, through
the legal amendments enforced, accrediting public sector (including the
Treasury) came to an end. The reforms implemented transformed the fundamental
interest of CBRT to shaping the expectations in the short run and this policy
paved the way for the strengthening of interest rate channel in monetary
transmission mechanism.
Another
aspect of the structural reform process implemented in Turkey was the
restructuring of public banks. Ineffectual branches were closed and their
employees were eliminated through distribution to other institutions, capital
adequacy ratios were increased and a new institution entitled Banking
Regulation and Supervision Agency (BDDK), which was responsible from supervision
of all the banks, was established. In a similar vein, reforms were implemented
in other public institutions for harmonization with European Union. These
reforms included the elimination of ineffectual units and positions in public
sector, reformation of agricultural support system, limitation of employment in
public institutions, ensuring the budget discipline in public sector and
introduction of new regulations with regard to accountability. Through the new
laws enforced and the amendments on the existing laws, changes were introduced
with regard to regulation and supervision of private sector with the aim of
increasing competition and economic efficiency.
In
the 5 years following the reforms, the annual economic growth rate increased to
7% on average and inflation decreased to single digit levels. A substantial
improvement was achieved in public sector indebtness and foreign debts of the
country. Airado et al. (2004, 1) note that
“the assessment of both private market actor program has been successful in
re-establishing macroeconomic stability, reducing the debt ratio and laying the
ground for a durable acceleration of growth in an environment of drastically
reduced inflation and much lower
New
conjuncture which emerged after the global crisis has lead central banks to
search for alternative policies. In this context, from the end of 2010, the
Turkish Central Bank (CBRT) has also designed and implemented a new monetary
policy framework. One of the important
lessons from the global crisis, while
focusing on the price stability, central banks should not overlook accumulated
risks and puffiness of asset prices in financial system. In this respect, the
view which central banks should give more importance to the opinion of the
financial stability such as the international platforms of the G-20 is
increasingly widespread. Also, in order to prevent financial crises the
importance of works which referring reaction with macro policies in recent
economic literature gradually gained value (Bianchi and Mendoza 2011, Jeanne
and Korinek, 2010).
Beside varying approaches about
changes in economic policies and central banks, the extraordinary global
economic conjuncture has also been a significant share with CBRT’snew
policy search. Damages caused by intense
crisis resulting from Lehman Brothers’ bankr in September 2008 and relating
policies applied by the developed countries have brought the
unusual dynamics with along. Although
five years after the crisis period, correction period of balance sheet which
started in developed countries after crisis is still continuing. Especially in
developed countries balance sheets of the public sector and the banking system
were entering in a negative mutual interaction and this leads to slowing the
recovery period. This circumstance also forces the limits of central banking.
From
mid-2009 while Turkey's economy has entered the process of rapid growth related
to domestic demand, recovery with main trade partners has been much slower. In
other words, during this period, it was observed that the internal and external
demand were in serious decomposition. Especially trend even became more evident
by increased short term capital flows after the second quantitative expansion
packs applied by the central banks of developed countries in 2010. While
capital inflows brought along excessive increase in value of Turkish Lira, at
the same time use of domestic credit has also been accelerated. As a result of
these developments, a serious distortion was observed in the balance of foreign
trade and the current account. During this period, because of the increase in
the current account deficit and capital inflows-quality degradation, economy
has become brittle against sudden changes in global risk trap. As of the last
quarter of 2010, it has been observed that almost the entire current account
deficit was financed by short-term capital and portfolio investments. While
looking at Turkey in particular with historical movements, it has been realized
that fluctuation in capital streams has been playing an important role on
macroeconomic stability.
Current
account deficit has been progressed in high levels due to the demographic
factors and other structural reasons, and this might affect the economy to be
become more brittle against fluctuations in capital movements. Limiting the
current account deficit to a great extent (cyclic portion of) and balancing the
external funding are being important in order to have macroeconomic stability
and sustainable growth. In 1994, 2001 and 2008 crises hard narrowing in
economic activity has always become along sudden cease in capital flows. This
observation signs that the durability of the economy against sudden changes in
global risk trap should be increased in a conjuncture with fairly short term
and fluctuating financing; at the same time it reveals the importance of a more
flexible approach to monetary policy (Kara, 2012).
5.
Conclusion
This
study evaluates the stance of monetary policy through the use of MCI and
discusses how interest rate and exchange rate can be taken into account at the
same time. In different policy analyses recommended, there is a common problem:
Despite the size of the sample, the weights of financial variables which are
used during the construction of MCI are fixed. However, change of portfolios of
the companies and households together with the business cycle and certain
economic events is an expected reaction. In order to overcome this problem, an
alternative methodology for the calculation of the weight of each asset is
recommended in this study. In order to determine the changes of weights over
time, the use of Kalman filter algorithm is the suggested. Financial markets
were included in the model with two variables through the consideration of the
traditional backward looking aggregate demand - aggregate supply model which
has been enhanced to include assets market: real interest rate and real
exchange rate. These variables were respectively represented through the use of
the gap between real and potential interest rate and real effective exchange
rate; long term values in relation with asset prices have been calculated
through the use of Hodrick-Prescott filtering technique. For the selection of
sample to constitute the basis for the study, the necessity to cover the
principal periods involving fundamental changes in public and monetary policy
was taken as basis and subsequently the period between January 1991 and
December 2013 was selected.
The
construction of MCI was divided into two phases: the first phase included the
estimation of aggregate demand and aggregate supply functions through the use
of Kalman filtering algorithm, whereas the second phase included the formation
of composite index by using coefficients which change over time. The method
used enables the determination of the impacts of unobservable factors likely to
influence output gap. Therefore, it is possible to observe the change of asset
price weight over time for each endogenous variable. MCI was established by
using the estimated weights for interest rate and exchange rate which vary over
time and the tendencies throughout the period were analyzed. Taken as a whole,
MCI can be said to fluctuate around (-1) throughout the period covered. On the
other hand, the increase in the volatility of the index during the periods
covering the years 2000-2003 and 2007-2009 attracts attention. The fundamental
finding of the empirical analysis carried out in this study is that the
relative significance of interest rate as the determinant of monetary
conditions is higher compared to exchange rate. Such a finding strengthens the
ability of the central bank with regard to the determination of policy stance.
In addition, the use MCI in the evaluation of economic and monetary conditions
is much more effective compared to the individual use of interest rate or
exchange rate as an independent variable.
Comments
Post a Comment