International Journal of Humanities and Social Science Review www.ijhssrnet.com Vol. 3 No. 2; May 2017
RELATIONSHIP CAUSALITY
BETWEEN BUDGET DEFICIT AND CURRENT: THE CASE OF TURKEY
Y. Koray Duman
University of Akdeniz, Department of Economics
07058, Antalya, Turkey
Abstract
This paper examines
the causal relationship between budget deficit and current account deficit, and
the direction of causality for Turkey
Keywords: Budget deficit, current account deficit, Twin deficit, Unit root, Causality, VAR model
JEL Classification codes: C22
Y. Koray
Duman is associate professor at
University of Akdeniz, faculty of Economics and Administrative Science,
department of Economics, Antalya, Turkey.
1 Introduction
After the oil crises of the 1970s, profoundly affecting the world economy in the USA, many of the country, in the early 1980s a large budget deficit and current account deficit to remain faced with the problem, has led to intensification of research on twins explains.Twin defined as an open relationship between budget deficit and current account deficit, particularly in attempts to maintain macroeconomic stability in the process of economic growth for the economies of developing countries has been an important policy issue. In recent years, especially in developing countries with current account and budget deficits increased rapidly in the debate and many studies have been conducted on this issue.
After the oil crises of the 1970s, profoundly affecting the world economy in the USA, many of the country, in the early 1980s a large budget deficit and current account deficit to remain faced with the problem, has led to intensification of research on twins explains.Twin defined as an open relationship between budget deficit and current account deficit, particularly in attempts to maintain macroeconomic stability in the process of economic growth for the economies of developing countries has been an important policy issue. In recent years, especially in developing countries with current account and budget deficits increased rapidly in the debate and many studies have been conducted on this issue.
In the literature on the twin hypothesis, the relationship between budget deficit and current account deficit, which the Keynesian approach and the relationship between these two there is an open advocate of Ricardian equivalence hypothesis is discussed in the context. Empirical studies focusing on the twin on the budget deficit and current account deficit (trade deficit) by examining the relationship between the Keynesian vision, or tried to test the validity of Ricardian Equivalence Hypothesis. However, studies of this relationship very different results were obtained in the presence and direction. The findings, examined the country, the research period, the method used and the data set that shows sensitivity. In addition, in countries where the macroeconomic conditions, especially the fragility of the country's economy and the most important elements in determining the direction of this relationship is prone to crisis.
Section 2
concerns relevant theoretical and empirical researches of …. Section 3 outlines
the econometric approach which includes data and variables, unit root tests,
causality tests and vector autoregressive model. Section 4 outlines empirical
results of our models. Finally, section 5 concludes.
2 Literature Review
In developed countries, and
particularly among developing countries in attempts to maintain macroeconomic
stability, rising budget deficits and current account deficits that accompany
it, the last two decades, the twin on the issue of increased interest and many
empirical studies done on this issue has been a pioneer.
In the literature, in order to explain the relationship between budget deficit and current account deficit are two basic approaches. The first approach, Keynesian Mundell-Fleming model, associated with (traditional) approach. According to the Keynesian approach, there is a positive relationship between budget deficit and current account deficit and budget deficit, foreign trade deficit of causality is correct. Twin clear hypothesis, that is, caused by an increase in budget deficits, current account deficit thesis is based on the Keynesian view. According to the Keynesian view, the occurrence of an increase in budget deficits, public sector borrowing and an increase in demand on the rise in domestic interest rates lead foreign interest rate. Flows of capital flows in response to higher interest rate domestically, leads to the appreciation of national currency (the relative prices of foreign goods is getting cheaper and increased propensity to import) and foreign trade deficit in the final of this situation (the current account deficit) results (Islam, 1998:122-123; Arize and Malindretos, 2008:4; Erdogan and Booth, 2009:138).
The second approach to the relationship between the two on the current account deficit could lead to budget deficits in the Keynesian approach, proposed by rejecting the hypothesis is based on the Ricardian equivalence hypothesis. For the first time in Barro (1974) proposed by the Ricardian equivalence hypothesis, according to a causal relationship between budget deficit and current account deficit is not. According to this approach, which reduces public revenues and public savings, and an expanding budget deficit, tax cuts, tax increases in coming years to meet an expected increase private savings, the same amount. Giving a positive response to changes in savings in the budget deficit, causing the trade deficit unchanged. Similarly, public expenditures to be financed by borrowing, the economic future for individuals to finance the public budget deficit, taxes will increase the waiting, so that individuals increase their savings to meet future tax increases.As a result, a change in the composition of public financing (debt or taxes) and real interest rates, aggregate demand, private spending, foreign exchange or has no effect on the current account balance. In other words, the causal relationship between the two on the lack of support of Ricardian equivalence hypothesis (Arize and Malindretos, 2008:4; Erdogan and Booth, 2009:138; Vyshnyak, 2000:16-17).
Applied studies in the literature, for the Keynesian or the economies of countries consists of studies to demonstrate the validity of Ricardian equivalence Hipotezleri'nin. Empirical studies in the context of a causal relationship between budget and trade deficit, likely to be classified according to four situation. The first case, the budget deficit, foreign trade deficit of causality is correct, Keynesian contact consists of supporting trials, the foreign trade deficit budget deficit in the second case to emphasize the existence of reverse causality works are included. According to the third situation, the trade deficit and budget deficit are two-way causality. These three cases in the literature, the existence of a causal relationship between budget and current account deficit to demonstrate the Keynesian response. In the fourth case, the absence of causality relationship between budget deficit and current account deficit, Ricardian equivalence hypothesis consists of supporting studies.
Initial studies in the 1980s often lived experience of a large budget deficit and current account deficit on the U.S. economy, while the current account deficit of causality budget deficits have reached the correct conclusions supporting the Keynesian view (Bahmani-Oskooee, 1989; Latif-Zaman, and Dacosta, 1990; Abell, 1990 ; Bachman, 1992). Bahmani-Oskooee (1989), the effect on the U.S. budget deficit, current account deficit for the period 1973-1985 using quarterly data examined, restores a contribution towards the budget deficit, current account deficit has reached the evidence supports the Keynesian view. Bernheim (1988), where the five countries in trade relations with the United States and open to the twin study examined the hypothesis that the traditional horizontal-sectional regression using the technique of budget deficits and current account deficits of at least one-third part be explained by the conclusion reached (Leachman and Francis, 2002:1122). Abell (1990), twin open the hypothesis that the
Leachman and Francis (2002), using both conventional cointegration and multiple co-integration analysis II. After World War I (1948-1992) for the
Twin clear whether the
current hypothesis for the economy of many countries outside of the U.S.
economy was also examined. This is one of the key studies Vamvoukas
(1999), Greece Greece Turkey , six
emerging market economy (Brazil ,
Mexico , Colombia , Czech
Republic and South Africa
Studies support the Keynesian view towards the budget deficit, foreign trade deficit, one-way causality, while highlighting some of the other studies, the current account deficit is evidence that budget deficits are the correct one-way causality. According to this approach, the current account deficit increases as a result of economic growth, and indirectly leads to budget deficits. This type of study is highly dependent on capital flows, especially the small and developing countries is valid samples. For example, Latin American and Southeast Asian countries, this situation can be observed in the presence of (Baharumshah et al, 2006).Baharumshah et al. (2006), Indonesia, Malaysia, the Philippines, and Thailand-for the VAR model, using the budget deficit, current account deficit (trade deficit) of the relationship. According to the results of the analysis, the long-term relationship between the budget deficit and current account deficit is valid, but the strength of this relationship varies by country. Results for
In some other studies in the literature, two-way causality between the two open-work consists of providing evidence to. Islam (1998),
In the last situation encountered in the literature, a causal relationship between budget and current account deficit, refused to support the work consists of Ricardian equivalence hypothesis.Miller and Russen (1989), Enders and Lee (1990) and Kaufman et al. (2002) studies, no causal relationship between budget deficits and current account deficit is not found.
The studies reached different findings for
3 Econometric Approach
The applied
data source is Turkey Statistics Institue (TÜİK) and Central Bank of Republic
of Turkey (TCMB). The data contains the series between 1994 and 2010 which have
the values for the first four months of each year, so each series has 68
values. We determined three variables for our models and they are the
proportion of budget deficit in gross domestic products (BD), the real exchange
rate (RER) and the proportion of current deficit in gross domestic products
(CD).
First of
all, we performed Augmented Dickey-Fuller tests for the series to check the presence
of unit roots. Then, we tested causality of the series and finally designed
vector autoregressive (VAR) model for the series.
3.1 Unit Root Tests
Stationary
series are necessary for analyzing them in time series models. So, before
modeling data, we checked presence of unit root for each series by using of
Augmented Dickey-Fuller (ADF) test. As seen in Table 1, 2 and 3; the series BD
has no unit root but the series RER and CD have unit root at original levels.
As a result of this, we took the first difference of the series RER and CD to
make them stationary. Correlograms of the series in Graph 2 support the results
of ADF tests in Table 1, 2 and 3. Consequently, we use the original series BD
and the first differences of the series RER and CD (D(RER) and D(CD)) in our
models. See Graph 1 for the time series lines of the all original and
differenced series. Also in Graph 1, it could be supported that the series we
used in model are all stationary.
|
Table 1: ADF tests for BD
|
|
|
||
|
Null Hypothesis: BD has a unit root
|
||||
|
Exogenous: Constant
|
||||
|
Lag Length: 0 (Automatic - based on AIC,
maxlag=10)
|
||||
|
|
|
|
t
|
Prob.*
|
|
|
|
|
-5.95733
|
0.000
|
|
Critical values:
|
1% level
|
-3.53159
|
|
|
|
|
|
5% level
|
-2.90552
|
|
|
|
|
10% level
|
-2.59026
|
|
|
Exogenous: Constant, Linear Trend
|
||||
|
Lag Length: 0 (Automatic - based on AIC,
maxlag=10)
|
||||
|
|
|
|
t
|
Prob.*
|
|
|
|
|
-7.11107
|
0.000
|
|
Critical values:
|
1% level
|
-4.10094
|
|
|
|
|
|
5% level
|
-3.47831
|
|
|
|
|
10% level
|
-3.16679
|
|
|
Exogenous: None
|
||||
|
Lag Length: 3 (Automatic - based on AIC,
maxlag=10)
|
||||
|
|
|
|
t
|
Prob.*
|
|
|
|
|
-1.00537
|
0.280
|
|
Critical values:
|
1% level
|
-2.6016
|
|
|
|
|
|
5% level
|
-1.94599
|
|
|
|
|
10% level
|
-1.6135
|
|
|
*MacKinnon (1996) one-sided p-values.
|
|
|||
[Table 2 about here]
|
Table 2: ADF tests for RER
|
|
|
||
|
Null Hypothesis: RER has a unit root
|
|
|||
|
Exogenous: Constant
|
|
|
||
|
Lag Length: 6 (Automatic - based on AIC,
maxlag=10)
|
||||
|
|
|
|
t
|
Prob.*
|
|
|
|
|
-0.44648
|
0.894
|
|
Critical values:
|
1% level
|
-3.5421
|
|
|
|
|
|
5% level
|
-2.91002
|
|
|
|
|
10% level
|
-2.59265
|
|
|
Exogenous: Constant, Linear Trend
|
|
|||
|
Lag Length: 1 (Automatic - based on AIC,
maxlag=10)
|
||||
|
|
|
|
t
|
Prob.*
|
|
|
|
|
-5.33098
|
0.000
|
|
Critical values:
|
1% level
|
-4.1032
|
|
|
|
|
|
5% level
|
-3.47937
|
|
|
|
|
10% level
|
-3.1674
|
|
|
Exogenous: None
|
|
|
|
|
|
Lag Length: 6 (Automatic - based on AIC,
maxlag=10)
|
||||
|
|
|
|
t
|
Prob.*
|
|
|
|
|
2.786252
|
0.999
|
|
Critical values:
|
1% level
|
-2.60342
|
|
|
|
|
|
5% level
|
-1.94625
|
|
|
|
|
10% level
|
-1.61335
|
|
|
Null Hypothesis: D(RER) has a unit root
|
|
|||
|
Exogenous: Constant
|
|
|
||
|
Lag Length: 5 (Automatic - based on AIC,
maxlag=10)
|
||||
|
|
|
|
t
|
Prob.*
|
|
|
|
|
-5.79907
|
0.000
|
|
Critical values:
|
1% level
|
-3.5421
|
|
|
|
|
|
5% level
|
-2.91002
|
|
|
|
|
10% level
|
-2.59265
|
|
|
Exogenous: Constant, Linear Trend
|
|
|||
|
Lag Length: 5 (Automatic - based on AIC,
maxlag=10)
|
||||
|
|
|
|
t
|
Prob.*
|
|
|
|
|
-5.74514
|
0.000
|
|
Critical values:
|
1% level
|
-4.11568
|
|
|
|
|
|
5% level
|
-3.48522
|
|
|
|
|
10% level
|
-3.17079
|
|
|
Exogenous: None
|
|
|
|
|
|
Lag Length: 3 (Automatic - based on AIC,
maxlag=10)
|
||||
|
|
|
|
t
|
Prob.*
|
|
|
|
|
-5.72745
|
0.000
|
|
Critical values:
|
1% level
|
-2.60219
|
|
|
|
|
|
5% level
|
-1.94607
|
|
|
|
|
10% level
|
-1.61345
|
|
|
*MacKinnon (1996) one-sided p-values.
|
|
|||
|
Table 3: ADF tests for CD
|
|
|
||
|
Null Hypothesis: CD has a unit root
|
|
|||
|
Exogenous: Constant
|
|
|
||
|
Lag Length: 8 (Automatic - based on AIC,
maxlag=10)
|
||||
|
|
|
|
t
|
Prob.*
|
|
|
|
|
-0.32406
|
0.915
|
|
Critical values:
|
1% level
|
-3.5461
|
|
|
|
|
|
5% level
|
-2.91173
|
|
|
|
|
10% level
|
-2.59355
|
|
|
Exogenous: Constant, Linear Trend
|
|
|||
|
Lag Length: 8 (Automatic - based on AIC,
maxlag=10)
|
||||
|
|
|
|
t
|
Prob.*
|
|
|
|
|
-1.95503
|
0.613
|
|
Critical values:
|
1% level
|
-4.1213
|
|
|
|
|
|
5% level
|
-3.48785
|
|
|
|
|
10% level
|
-3.17231
|
|
|
Exogenous: None
|
|
|
|
|
|
Lag Length: 8 (Automatic - based on AIC,
maxlag=10)
|
||||
|
|
|
|
t
|
Prob.*
|
|
|
|
|
0.507667
|
0.822
|
|
Critical values:
|
1% level
|
-2.60475
|
|
|
|
|
|
5% level
|
-1.94645
|
|
|
|
|
10% level
|
-1.61324
|
|
|
Null Hypothesis: D(CD) has a unit root
|
|
|||
|
Exogenous: Constant
|
|
|
||
|
Lag Length: 7 (Automatic - based on AIC,
maxlag=10)
|
||||
|
|
|
|
t
|
Prob.*
|
|
|
|
|
-3.06966
|
0.034
|
|
Critical values:
|
1% level
|
-3.5461
|
|
|
|
|
|
5% level
|
-2.91173
|
|
|
|
|
10% level
|
-2.59355
|
|
|
Exogenous: Constant, Linear Trend
|
|
|||
|
Lag Length: 7 (Automatic - based on AIC,
maxlag=10)
|
||||
|
|
|
|
t
|
Prob.*
|
|
|
|
|
-3.0612
|
0.125
|
|
Critical values:
|
1% level
|
-4.1213
|
|
|
|
|
|
5% level
|
-3.48785
|
|
|
|
|
10% level
|
-3.17231
|
|
|
Exogenous: None
|
|
|
|
|
|
Lag Length: 7 (Automatic - based on AIC,
maxlag=10)
|
||||
|
|
|
|
t
|
Prob.*
|
|
|
|
|
-2.70283
|
0.008
|
|
Critical values:
|
1% level
|
-2.60475
|
|
|
|
|
|
5% level
|
-1.94645
|
|
|
|
|
10% level
|
-1.61324
|
|
|
*MacKinnon (1996) one-sided p-values.
|
|
|||
Graph 1: Time Series
Graph for Series
Graph 2: Correlograms for Series
Correlogram
for BD
Correlogram
for RER
Correlogram
for D(RER)
Correlogram
for CD
Graph 2: Continued
Correlogram
for D(CD)
3.2 Granger Causality Test and Variable Order
We performed Granger
Causality test to get pairwise causality between the series and put an order
the series for VAR model.
|
Table 4: Granger Causality Tests
|
|
|
|
|
Null Hypothesis:
|
Obs
|
F-Statistic
|
Prob.
|
|
|
|
|
|
|
D(RER) does not Granger Cause BD
|
63
|
1.15825
|
0.340
|
|
BD does not Granger Cause D(RER)
|
|
0.54084
|
0.706
|
|
|
|
|
|
|
D(CD) does not Granger Cause BD
|
63
|
0.68724
|
0.604
|
|
BD does not Granger Cause D(CD)
|
|
2.7606
|
0.037
|
|
|
|
|
|
|
D(CD) does not Granger Cause D(RER)
|
63
|
1.37458
|
0.255
|
|
D(RER) does not Granger Cause D(CD)
|
|
2.17403
|
0.084
|
|
Lags: 4
|
|
|
|
According to resulst of
Granger causality tests, as seen in Table 4, there is no pairwise causality
releationship between the series BD and D(RER). However, there is one way
causality from BD to D(CD) at .05 level and also there is one way causality
from D(RER) to D(CD) at .10 level. In other words, BD causes D(CD) and D(RER)
causes D(CD) in respect of Granger test. As a result, we put order the series
as BD→D(RER)→D(CD).
3.3 VAR model
We selected lag order 4 for
our VAR model according to Akaike information criteria (AIC). And also,
sequential modified LR test statistic determines lag order 4. Shortly, we use
VAR(4) model for our series.
|
Table 5: VAR Serial Correlation LM Tests
|
||
|
Included observations: 63
|
|
|
|
|
|
|
|
Lags
|
LM-Stat
|
Prob
|
|
|
|
|
|
1
|
6.141065
|
0.726
|
|
2
|
12.48595
|
0.187
|
|
3
|
9.388545
|
0.402
|
|
4
|
7.953261
|
0.539
|
|
5
|
6.835709
|
0.654
|
|
6
|
15.47015
|
0.079
|
|
7
|
9.877044
|
0.361
|
|
8
|
6.973632
|
0.640
|
|
9
|
7.88739
|
0.546
|
|
Probs from chi-square with 9 df.
|
|
|
|
Table 6: VAR White Heteroskedasticity
Test
|
||
|
Chi-sq
|
df
|
Prob.
|
|
|
|
|
|
154.6468
|
144
|
0.257
|
It can be seen in Table 5
that there is no autocarrelation and according to Table 6, there is no heteroskedasticity
problem in our model. In addition, inverse roots of the model are welldone as
seen in Graph 3.
4 Empirical Results
Graph 4 shows
impulse-response functions of the series. Impulse-response graph show that
resulting one unit shock in the one of the series is responsed by the other
series.
Graph 4: Impolse- Response
functions
The response of D(RER) is
positive to one unit shock in BD. After the period 3, this effect of D(RER) is
disappearing. The response of D(CD) is negative to one unit shock in BD. After
period 3, this effect of D(RER) is becoming positive and after period 6,
disappearing. The response of D(CD) is negative to one unit shock in D(RER). After
period 4, this effect of D(CD) is becoming positive and after period 6, disappearing.
In Table 7, 8 and 9; it can
be seen the variance decomposition of BD, D(RER) and D(CD) respectively.
|
Table 7: Variance Decomposition of BD
|
|
|||
|
Period
|
S.E.
|
BD
|
D(RER)
|
D(CD)
|
|
|
|
|
|
|
|
1
|
0.037225
|
100.0000
|
0.000000
|
0.000000
|
|
2
|
0.038733
|
97.29198
|
2.691735
|
0.016286
|
|
3
|
0.039373
|
95.51990
|
4.448579
|
0.031521
|
|
4
|
0.040385
|
94.07299
|
4.772825
|
1.154187
|
|
5
|
0.04189
|
91.14064
|
4.442805
|
4.416558
|
|
6
|
0.042077
|
91.17857
|
4.410331
|
4.411102
|
|
7
|
0.042209
|
91.15295
|
4.392257
|
4.454793
|
|
8
|
0.042443
|
91.19037
|
4.357191
|
4.452441
|
|
9
|
0.042732
|
91.09920
|
4.431983
|
4.468822
|
|
10
|
0.042816
|
91.01810
|
4.507097
|
4.474803
|
|
Table 8: Variance Decomposition of
D(RER)
|
||||
|
Period
|
S.E.
|
BD
|
D(RER)
|
D(CD)
|
|
|
|
|
|
|
|
1
|
8.196719
|
0.910846
|
99.08915
|
0.000000
|
|
2
|
8.566847
|
4.057401
|
90.75210
|
5.190496
|
|
3
|
8.939959
|
3.778140
|
91.34376
|
4.878099
|
|
4
|
9.182003
|
4.014027
|
87.09309
|
8.892887
|
|
5
|
9.425406
|
3.947437
|
85.69832
|
10.35424
|
|
6
|
9.458187
|
4.149974
|
85.28432
|
10.56571
|
|
7
|
9.519031
|
4.193345
|
84.89711
|
10.90955
|
|
8
|
9.579058
|
4.805488
|
83.84853
|
11.34598
|
|
9
|
9.615649
|
4.792540
|
83.94732
|
11.26014
|
|
10
|
9.627357
|
4.890412
|
83.74328
|
11.36631
|
|
Table 9: Variance Decomposition of D(CD)
|
||||
|
Period
|
S.E.
|
BD
|
D(RER)
|
D(CD)
|
|
|
|
|
|
|
|
1
|
3.62E-05
|
0.328247
|
7.213868
|
92.45788
|
|
2
|
3.95E-05
|
2.348760
|
6.056786
|
91.59445
|
|
3
|
3.99E-05
|
3.081337
|
6.295379
|
90.62328
|
|
4
|
4.16E-05
|
9.318532
|
7.343478
|
83.33799
|
|
5
|
4.44E-05
|
9.538785
|
14.78685
|
75.67436
|
|
6
|
4.49E-05
|
9.739358
|
14.84504
|
75.41560
|
|
7
|
4.51E-05
|
9.780662
|
15.03516
|
75.18418
|
|
8
|
4.51E-05
|
9.761460
|
15.00402
|
75.23452
|
|
9
|
4.57E-05
|
9.574942
|
16.17657
|
74.24849
|
|
10
|
4.58E-05
|
9.575270
|
16.10736
|
74.31737
|
According to Table 7, while
the series BD is first effected by own lagged values amoung ten periods, the
effectives of the other series are not considerable. The effective of D(CD)
picks up after period 5 and after that it has equal effective with D(RER) in
the other periods.
In Table 8, the series
D(RER) is effected by own lagged values amoung ten periods. The series D(CD) is
more effective than the series BD. The effect of D(CD) is higher than 10
percent after period 5 while the effect of BD is almost same for all ten
periods.
In Table 9, the series
D(CD) is effected by own lagged values amoung ten periods. The series D(RER) is
more effective than the series BD. The effect of D(RER) is higher than 14
percent after period 5 and growing up while the effect of BD is picks up after
period 4 and goes on equal amoung 10 period.
5 Conlusion
While the Keynesian approach
emphasizes that internal deficit would co-exist with external deficit and would
be equal to it; the Ricardian Equivilance that came to the fore with Barro’s
1974 paper, argues that above mentioned Keynesian result may not be obtained
due to the Ricardian Equivilance that came to the fore with Barro’s 1974 paper,
argues that above mentioned Keynesian result may not be obtained due to an alteration in household saving behaviour.
In this study,the relationship
between budget deficits and current account deficits in Turkey for 1994-2010 periodis examined with annual
data by using recently developed time series econometric tecniques such as
thecointegration analysis and Granger Causality Test. In cointegration
analysis, it is found that there is along run relationship between budget
deficits and current account deficits. This conclusion supports theConventional
Keynesian Theory. Moreover, according to Granger causality tests results, the
causality runs from budget deficits to current account deficit
KAYNAKÇA
Kaynakça
Abell, John D.
(1990), Twin Deficits During the 1980s: An Empirical Investigation, Journal of
Macroeconomics, 12 (1), 81-96.
Akbostancı, Elif,
Tunç, Gül İpek (2002), Turkish Twin Deficits: An Error Correction Model of
Trade Balance, Economic Research Center Working Papers in Economics 01/06.
Aksu, Hayati ve
Başar, Selim (2005), İkiz Açıklar Hipotezi’nin Türkiye Açısından Araştırılması,
İktisat İşletme ve Finans, Yıl.20, Sayı 234, 109-114.
Alkswani, Mamdouh
A. (2000), The Twin Deficits Phenomenon in Petroleum Economy: Evidence from
Saudi Arabia, Seventh Annual Conference, Economic Research Forum (ERF), 26-29
October, Amman, Jordan.
Anoruo, Emmanuel ve
Ramchander, Sanjay (1998), Current Account and Fiscal Deficits: Evidence from
Five Developing Economies of Asia, Journal of Asian Economics, 9 (3), 487-501.
Arize, Augustine C.
ve Malindretos, John (2008), Dynamic Linkages and Granger Causality Between
Trade and Budget Deficits: Evidence from Africa, African Journal of Accounting,
Economics, Finance and Banking Research, Vol.2, No.2, 1-19.
Asteriou, Dimitrios
(2006), Applied Econometrics: A Modern Approach Using Eviews and Microfit, Pall
Grave MacMillan, New York.
Ay, Ahmet, Karaçor,
Zeynep, Mucuk, Mehmet ve Erdoğan, Savaş (2004), Bütçe Açığı-Cari İşlemler Açığı
Arasındaki İlişki: Türkiye Örneği (1992-2003), Selçuk Üniversitesi SBE Dergisi,
12, 75-82.
Bachman, Daniel
David (1992), Why Is the US Current Account Deficit So Large? Evidence from
Vector Autoregressions, Southern Economic Journal, 59 (2), 232-240.
Baharumshah, Ahmad
Zubaidi, Lau, Evan, Khalid Ahmed M. (2006), “Testing Twin Deficits Hypothesis
Using VARs and Variance Decomposition”, Journal of the Asia Pacific Economy,
Vol.11, No.3, 331-354.
Bahmani-Oskooee,
Mohsen (1989), Effects of the U.S. Government Budget on Its Current Account: An
Empirical Inquiry, Quarterly Review of Economics and Business, 29 (4), 76-91.
Bilgili, Faik ve
Bilgili, Emine (1998), Bütçe Açığının Cari İşlemler Dengesi Üzerindeki
Etkileri: Teori ve Uygulama, İktisat İşletme ve Finans, Yıl.13, Sayı.146, 4-16.
Chowdhury, Khorshed
ve Saleh, Ali Salman (2007), Testing the Keynesian Proposition of Twin Deficits
in the Presence of Trade Liberalisation: Evidence from Sri Lanka, University of
Wollongong Economics Working Paper Series, WP 07-09, 1-33.
Çelik, Sadullah ve
Deniz, Pınar (2009), An Empirical Investigation of Twin Deficit Hypothesis for
Six Emerging Countries, 1. Uluslararası Avrasya İşletme ve Ekonomi Kurumu
Konferansı, Kadir Has Üniversitesi, İstanbul, Türkiye.
Dickey, David A.ve
Fuller, Wayne A. (1979), Distribution of The Estimators for Autoregressive Time
Series with A Unit Root, Journal of the American Statistical Association, 74
(366), 427-431.
Dickey, David A.ve
Fuller, Wayne A. (1981), Likelihood Ratio Statistics for Autoregressive Time
Series with A Unit Root, Econometrica, 49, 1057-1072.
Enders, Walter ve
Lee, Bong-Soo (1990), Current Account and Budget Deficits: Twins or Distant
Cousins, Review of Economics and Statistics, 72 (3), 373-381.
Erdinç, Zeynep
(2008), İkiz Açıklar Hipotezinin Türkiye’de 1950-2005 Yılları Arasında
Eşbütünleşme Analizi ve Granger Nedensellik Testi ile İncelenmesi, Anadolu
Üniversitesi Sosyal Bilimler Dergisi, Cilt:8, Sayı:1, 209-222.
Erdoğan, Seyfettin
ve Bozkurt, Hilal (2009), Türkiye’de Cari Açığın Belirleyicileri: Mgarch
Modelleri ile Bir İnceleme, Maliye Finans Yazıları, Yıl:23, Sayı:84, 135-172.
Gomez, Victor ve
Maravall, Agustin (1994), Estimation, Prediction and Interpolation for
Nonstationary Series with the Kalman Fitler, Journal of the American
Statistical Association, 89 (426), 611-624.
Gomez, Victor ve
Maravall, Agustin (1996), Programs TRAMO (Time Series Regression with ARIMA
Noise, Missing Obsarvations and Outliers) and SEATS (Signal Extraction in ARIMA
Time Series)”, Instructions for User, Banca de Espana Working Paper, No. 9628.
Granger, Clive W.J.
(1969), Investigating Casual Relations by Econometric Models and Cross-Spectral
Methods, Econometrica, 37, 424-438.
Islam, M. Faizul
(1998), Brazil’s Twin Deficits: An Empirical Examination, Atlantic Economic
Journal, 26 (2), 121-28.
Kaufman, Sylvia,
Scharler, Johann, Winckler, Georg (2002), The Austrian Current Account Deficit:
Driven by Twin Deficits or by Intertemporal Expenditure Allocation?, Empirical
Economics, 27, 529-542.
Kearney, Colm ve
Monadjemi, Mehdi (1990), Fiscal Policy and Current Account Performance:
International Evidence on the Twin Deficits, Journal of Macroeconomics, 12 (2),
197-219.
Kwiatkowski, Denis,
Phillips, Peter, Schmidt, Peter ve Shin, Yongcheol (1992), Testing the Null
Hypothesis of Stationarity Against the Alternative of a Unit Root, Journal of
Econometrics, 54, 159-178.
Latif-Zaman, Nazma
ve DaCosta, Maria N. (1990), The Budget Deficit and Trade Deficit: Insights
into This Relationship, Eastern Economic Journal, 16 (4), 349-354.
Leachman, Lori ve
Francis, Bill (2002), Twin Deficits: Apparition or Reality?, Applied Economics,
34 (9), 1121-1132.
MacKinnon, James G.
(1991), Critical Values for Cointegration Tests, Long Run Economic
Relationships: Reading in Cointegration içinde, Engle, Robert F. ve Granger,
Clive W.J., Oxford University Press, Oxford, 267-276.
Miller, Stephen M.
ve Russek, Frank S. (1989), Are Twin Deficits Really Related?, Contemporary
Policy Issues, 7 (4), 91-115.
Pahlavani, Mosayeb
ve Saleh, Ali Salman (2009), Budget Deficits and Current Account Deficits in
the Philippines: A Causal Relationship?, American Journal of Applied Sciences,
6 (8), 1515-1520.
Phillips, Peter ve
Perron, Pierre (1988), Testing for a Unit Root in Time Series Regression,
Biometrica, 75, 335-346.
Piersanti, Giovanni
(2000), Current Account Dynamics and Expected Future Budget Deficits: Some
International Evidenve, Journal of International Money and Finance, 19 (2),
255-271.
Sever, Erşan,
Demir, Murat (2007), Türkiye’de Bütçe Açığı ile Cari Açık Arasındaki
İlişkilerin VAR Analizi ile İncelenmesi, Eskişehir Osmangazi Üniversitesi İİBF
Dergisi, 2 (1), 47-63.
Utkulu, Utku
(2003), Türkiye’de Bütçe Açıkları ve Dış Ticaret Açıkları Gerçekten İkiz mi?
Koentegrasyon ve Nedensellik Bulguları, Dokuz Eylül Üniversitesi İİBF Dergisi,
18 (1), 45-61.
Vyshnyak, Olga
(2000), Twin Deficit Hypothesis: The Case of Ukraine, s.1-54,
http://www.kse.org.ua/uploads/file/library/2000/Vyshnyak.pdf , (Erişim tarihi:
01.11.2010).
Vamvoukas, George
A. (1999), The Twin Deficits Phenomenon: Evidence from
Greece, Applied
Economics, 31 (9), 1093-1100.
Zengin, Ahmet
(2000), İkiz Açıklar Hipotezi (Türkiye Uygulaması), Ekonomik Yaklaşım, Gazi
Üniversitesi, C.2, S.35, 37-67.
Hiç yorum yok:
Yorum Gönder