Content uploaded by Coşkun Karaca
Author content
All content in this area was uploaded by Coşkun Karaca on Jan 19, 2016
Content may be subject to copyright.
Available via license: CC BY
Content may be subject to copyright.
International Journal of Economics and Financial Issues
Vol. 2, No. 3, 2012, pp.281-295
ISSN: 2146-4138
www.econjournals.com
Inflation and Corruption Relationship: Evidence from Panel Data in
Developed and Developing Countries
Haşim Akça
Çukurova University, Faculty of Economic and Administrative Sciences,
Department of Finance, Adana, Turkey. Email: hakca@yahoo.com
Ahmet Yilmaz Ata
Gaziantep University, Faculty of Economic and Administrative Sciences,
Department of Economics, Gaziantep, Turkey. Email: ayata@gantep.edu.tr
Coşkun Karaca
Cumhuriyet University, Faculty of Economic and Administrative Sciences,
Department of Finance, Erzurum, Turkey. Email: coskunkaraca@cumhuriyet.edu.tr
ABSTRACT: Corruption, which is defined as the illegal and benefit-oriented usage of public power,
is a fact that has an impact on the macro-economic performance of economy in the scope of cause and
effect. Within this framework, there is a strong cause and effect interaction between inflation, an
important economic parameter, and corruption. Inflation is defined as not only a financial factor
results in corruption but also an economic problem results from corruption. With this particular study,
the relationship between inflation and corruption was tried to be tested one-way. In this context, the
impact of inflation, growth, trade gap, the quality of legislation, the efficacy of government, political
stability and responsibility variables on corruption was tested through panel data method concerning to
the 2002-2010 period of totally 97 countries from three different income-level group. It was found as a
result of the empirical data that the inflation has a statistically significant and positive effect on
corruption in all these 97 countries from three different income-level groups.
Keywords: Inflation; Corruption; Panel Data
JEL Classifications: D02; D40; D72; D73; D82
1. Introduction
The case of corruption, which is defined as “the misuse of public power for private purposes”
(Gray and Kaufman, 1998: 7; Rose-Ackerman, 1999: 91; Bardhan, 1997: 1321; Klitgaard, 1988: 23;
Lambsdorff, 2007: 16), is a versatile concept which is in various forms and functions and which has
many reasons and results (Aidt, 2003: 632).
Quite a few economic, political and social dynamics can be reasons for the corruption activities
to appear. However, the dynamics that result in corruption can be shown in a simple model as follows
(Klitgaard, 1998: 75):
In this equation, states the level of corruption; states the power of monopoly; the
judicial discretion and states accountability. Hereunder; the level of corruption is defined by the
power of monopoly, the judicial discretion and accountability. While having the power of monopoly
and judicial discretion increase the level of corruption, the accountability decreases it. The power of
monopoly and judicial discretion is more commonly seen in economies in which public interventions
are much. Accordingly, it can be said that corruption is more prominent in societies which are headed
by interfering governmental structures although it is seen more or less in all economies depending on
the factors peculiar to the countries.
Corruption is both a moral problem in cultural and individual sense and an important problem in
economic and political life. In this sense, corruption is a sociological fact in terms of ethics and it is
International Journal of Economics and Financial Issues, Vol. 2, No. 3, 2012, pp.281-295
28
2
also dealt with as an economic fact due to its effects on social welfare and development (Andving et
al., 2000: 9; Luo, 2004:121).
The subject of corruption was under investigation of sociology, political science, history, public
administration and the science of law until 1980s inclusively. The economic analysis of corruption
began by 1980s and later it provided inspiration for more extensive studies. (Abed and Gupta, 2002:3).
In these studies, especially the economic results of corruption were focused on, but economic reasons
of corruption were also studied.
Although there were results different from each other, the common point that was reached at the
end of the studies about the economic results of corruption was that the effect of corruption on
economy was negative. The biggest damage of corruption on economy is the decrease in investments
and deceleration in the economic growth and development. Nevertheless, it is another approach that
corruption affects the distribution of the existing resources in economy and their effective usage
negatively and it causes inflation and inequality in the distribution of income (Al-Marhubi, 2000: 199).
The analysis about corruption and its economic reasons showed that many factors are effective
on the relationship between these two facts. Inadequacy of capacity that occurs when the supply
cannot meet the demand is one of the primary factors which results in corruption (Adaman et al.,
2001: 18). The role and the policies of the state, poverty, the structure of tax system, inequality in the
distribution of income, commercial limitations, inflation, low wages, and the competition power of
economy, index of openness, unrecorded economy and low employment are the other factors which
can cause corruption (Akcay, 2001: 44-45).
When both factors result in corruption and the effects of corruption are considered, the terms of
“inflation” and “corruption” have become the basis of an important research field. In the literature of
economics, there are not enough recent studies on the relationship between corruption and inflation
even though plenty of researches were done on corruption and its economic reasons. Particularly, there
are scarcely any applied studies on this subject in the literature. Most of these studies are about the
effect of corruption on inflation. In this particular study, it was firstly aimed to contribute to the related
literature on the “inflation-corruption” topic. In this scope, the effect of inflation on corruption in 97
countries from low, middle and high income levels in the period between 2002 and 2010 will be
analyzed through panel data prediction method considering the relationship between the variables of
corruption, inflation, growth, and legislation quality, the efficacy of government, political balance and
responsibility.
2. The Relationship Between Corruption and Inflation
2.1. Literature Review
Although corruption is a fact that has been seen in nearly all societies since antique ages, the
economic reasons and results of corruption could not be investigated empirically because it was
difficult to measure. However, the initiator studies about the measurement of corruption and the ability
to reach the data sources thanks to the increase in the databases enhanced the number of empirical
studies carried out on “corruption-inflation”. These studies were mostly done in order to investigate
the relationship between one or several components that constitute the reasons and results and
corruption (Ata, 2009: 268).
It is seen that the method which has generally been used in applied studies is multi-country
estimates depending on cross-sectional and/or panel data. It is also observed that the corruption
perceptions indexes which are calculated by highly reliable international institutions have been used
quite often. Within the last quarter century period, one of the research fields in the scope of “the
economic analysis of corruption” has been realized about the relationship of “inflation-corruption”.
Even though there are plenty of recent studies on corruption and its economic reasons, the
studies which focus on the relationship between inflation and corruption were not at the desired level
(Piplica, 2011: 471). In the limited number of these studies, however; it was found out that there is a
strong relationship between corruption and inflation. There are various arguments about the direction
of this interaction in the literature, though.
While some of the studies expressed that inflation causes corruptions, some of them claimed
that this interaction was in the opposite direction, that is, corruptions cause inflation.
Braun and Di Tella (2004) tested the interaction between inflation and corruption in 75
countries for the years between 1982 and 1994 through panel data with least squares method. The
Inflation and Corruption Relationship: Evidence from Panel Data in Developed and
Developing Countries
283
corruption index of ICRG was used as the dependent variable in the model. The variable of inflation,
import/GDP and the index of political rights were used as independent variables. The researchers
reached in their study that the change (the increase) in the rate of inflation caused a positive and
statistically significant effect on corruption. That is, some findings which showed that important
changes in the raise of prices increased corruption were obtained.
Paldam (2002) dealt with the reasons of corruption in the economic and cultural framework in
his study. Paldam investigated the factors that revealed corruption in 100 developed and developing
countries in the scope of economic and cultural models by using the corruption perceptions indexes of
Transparency International for 1999 and tested these models with least squares method and cross-
sectional analysis. Paldam took factors such as economic development, growth, inflation, economic
freedoms and unfair distribution of income into the extent of this analysis. Inflation can effect
corruption for a short period of time like 5 or 10 years. Hereunder, the increases in inflation raise
corruption.
Getz and Volkema (2001) investigated the interaction between corruption and economic and
cultural factors through least squares method and cross-sectional analysis. In the end of the study, they
included the economic development, the economic ambiguity and the bureaucratic structure as part of
economic conditions into the investigation scope of the study. Economic ambiguity is defined as the
increase in general level of prices. According to the findings obtained, it was concluded that
corruption goes up when the economic ambiguity, in other words, inflation increase.
Ata (2009) handled the factors resulting in corruption in his study in terms of economic and
social factors and it was found out that inflation causes corruption by taking the average four-year
(2004-2007) values of 25 European Union member countries and analyzing through cross-sectional
data analysis method.
Similarly Tosun (2002) analyzed the economic factors resulting in corruption for the 1982-1995
periods of 44 countries through panel data method and he presented that there was a statistically
significant and positive relationship between inflation and corruption.
On the other hand, Al-Marhubi (2000), who provided a significant contribution to the literature
about the relationship between corruption and inflation, claim that corruption increase inflation. The
writer tested the relationship between corruption and inflation in his study in which he used cross-
sectional data of 41 countries. The average annual inflation values of 41 countries for the years
between 1980 and 1995 were taken as the dependent variable. Corruption indexes prepared by
Transparency International and Business International were used as the data about corruptions. In his
analysis, the writer found a positive relationship between corruptions and inflation. In other words,
high inflation was observed in economies in which corruption was seen intensively (Al-Marhubi,
2000: 201). Similarly, Abed and Davoodi (2002), Smith-Hillman (2007), Samimi et al. (2012), Piplica
(2011), Ekpo (1985), Bahmani-Oskooee and Nasir (2002), Oweye and Bendarfdaf (1996) investigated
the effect of corruption on general level of prices in their studies and concluded that corruption
increased the prices.
Consequently, it can be said that there is a positive relationship between inflation and corruption
according to the findings obtained from empirical studies in which corruption was taken as the
dependent variable and inflation as the independent variable, or vice versa. These studies were shown
in Table 1.
Table 1. Studies on Inflation-Corruption nexus
Corruption Inflation
Braun and Di Tella (2004) Dep.variable -
Paldam (2002) Dep.variable -
Getz and Volkema (2001 Dep.variable -
Ata (2009) Dep.variable -
Tosun (2002) Dep.variable -
Al- Marhubi (2000) - Indep. variable
Abed and Davoodi (2002) - Indep. variable
Samimi et al. (2012) - Indep. variable
Piplica (2011) - Indep. variable
Vindelyn and Smith (2007) - Indep. variable
International Journal of Economics and Financial Issues, Vol. 2, No. 3, 2012, pp.281-295
284
2.2. Theoretical Framework
According to the findings of scientific studies, there is a significant and positive relationship
between the change in the inflation rate and corruption. That is, a significant change in prices has an
increasing impact on corruption (Braun and Di Tella, 2004: 79). We can talk about many reasons that
cause the interaction between inflation and corruption. These reasons can be explained as follows:
It is commonly believed by the public that inflation, which can be defined as the increase in
general level of prices due to the raise in money supply, causes moral erosion (Paldam, 2002: 221) and
creates more opportunities for illegal and unethical behaviours such as jugglery or cheating (Braun and
Di Tella, 2004: 80). Accordingly, the increase in inflation and fast change also results in corruptions.
According to this view, the countries which have low inflation rates should be also the ones which
have low corruption rates. The inflation and corruption index data of ten countries which has the
lowest corruption rates in 2009 that were presented in Figure 1 shows a characteristic verifying this
consideration. According to the data, the countries in which the lowest corruption rates are observed
are the ones which also have very low inflation rates (single-digit numbers).
Figure 1. The relationship of Corruption-Inflation (2009)
Source: Transparency International (2009); Corruption Perceptions Index, 2009 and World
Economic Forum (2009); Global Competitiveness, Report 2009.
High inflation is an agent which brings about revenue loss of individuals and groups, decay of
the income distribution, increase in rent-seeking activities and emerge of ambiguity in economy (Al-
Marhubi, 2000: 199; Husted, 1999: 340; Haider et al., 2011: 3).
Within this scope, inflation affects the purchasing power of individuals and groups negatively
by lowering the real wage level (Tosun, 2002: 81). Individuals and groups must fulfil their needs
although their purchasing power decreases. This might result in corruptions as individuals and groups
can look for illegal methods (Al-Marhubi, 2000: 200). Along the same line, inflation that also causes
the decrease in the value of the money reduces the real incomes of civil servants employed in the
public sector, spoils the distribution of income and supports the large capital owners. The imbalance in
the distribution of income naturally stimulates corruption behaviours (Husted, 1999: 342; You and
Khagram, 2005: 5; Gupta et al., 1998: 21).
Another important reason of the increase in inflation to cause corruption is about the high
expenditure of governments (Haider et al., 2011: 6). The high level of current and investment
expenditures also brings along financial difficulties. The expenditures increasing due to the populist
behaviours of political authorities and not being able to raise the tax revenues in order not avoid the
reaction of voters make the governments run up debts and mintage. (Samimi et al., 2012: 392). Being
ungovernance starting from money authorities and spreading to other public institutions result in
Inflation and Corruption Relationship: Evidence from Panel Data in Developed and
Developing Countries
285
lubberliness of bureaucracy and this leads private sector to illegal behaviours and civil servants who
are defeated by inflation to corruption so as to meet their recurring expenditure level. Briefly, high
inflation might affect the economic degeneration by determining the usage of public resources,
increasing rent seeking and lobbying activities (Rahmani and Yousefi, 2009: 3).
Furthermore, inflation’s increasing rent-seeking activities and effect on spreading corruption are
seen less common in developed countries which has political stability and in which the quality of
legislation and the rule of law are dominant. Typical characteristics of industrialized countries which
have high income can be described with a low level of bureaucratic corruption which is provided by
low inflation, strong growth and better governance (Huang and Wei, 2003: 3). On the other hand, less
developed countries with low income face with many difficulties such as governance with weak
economic performance, high level of corruption, high inflation caused by seigniorage addiction in
order to finance the public expenditures and stagnant growth. Distorted macroeconomic policies that
these countries follow instigate high inflation by causing large budget and current account deficits
(Haider et al., 2011: 8). Weak institutions that are under pressure in a period like which high inflation
is seen, property rights that cannot conserve investors and political instability has created a suitable
environment for corruption (Samimi et al., 2012: 392).
Ambiguity in economic life that is caused by inflation might result in unfulfilled functions of
price
1
(Tosun, 2002: 81). The relationship between corruption and inflation has focused most on the
function of “the transfer of knowledge” in terms of the functions of price. Accordingly, the price
cannot fulfil the function of “the transfer of knowledge” due to the rapid change in inflation and this
causes increase in corruptions. The existence of high and changeable inflation raises the ambiguity
about the future prices. A situation like this will make the supervision of individual behaviours more
costly. (Braun and Di Tella, 2004: 79-80). Hereunder, it is quite difficult and costly to take the prices
that salesmen report under control due to the continuous change in an environment like this. In other
words, it is meant that public officials can show the invoice amount more than normal and sellers can
show it less than normal in environments in which inflation is high and changes rapidly (Tosun, 2002:
82). This provides suitable conditions for illegal and unjustified benefits. Inflation might contribute the
spread of corruption by increasing the thoughts and tendencies such as looking for speculative earning,
engrossing and hitting the jackpot.
Getz and Volkema (2001: 12) think that the existence of economic depressions caused by
inflation, unemployment and recession results in an increase in corruptions because appearance of
problems like these in economy is an important factor which generates a loss of trust towards the
central authority. In this context, the existence of inflation raises corruptions as it increases the
ambiguity in economic life and lack of confidence (Paldam, 2002: 222). The fluctuation of inflation
rates limits the prices’ function of “the transfer of knowledge”. In this case, price revisions between
time of procurement and of delivery in public procurement costs become a current issue. While a part
of revisions are based on legitimate price escalation, another part of them might cause corruptions
(Celen, 2007: 94).
Inflation could influence corruptions implicitly, too. The increase of inflation can lower the
investments and economic growth and it can make the level of corruption higher due to these indirect
effects (Braun and Di Tella, 2004: 80). Invariably, inflation pushes disparity of income distribution in
society up and this might result in the enlargement of corruptions. (Paldam, 2002: 222).
Conversely, the relationship between inflation and corruption could cause an effect bilaterally.
The increase in corruptions both makes public incomes decrease (capital stocks escape to other
countries and this makes the resources which can be taxed and therefore tax incomes decrease) and
public expenditures increase (in the economies in which corruption is widespread, the governments
1
Hayek described the functions of price in his study named “The Uses of Knowledge in Society” (1945) as
follows. According to this, price has got three functions. The first and the most important of them is “the transfer
of knowledge”. For example, people realize the fact that they should not waste energy quickly due to the
increasing prices of energy. These functions of price help an important function like the coordination of
economic activities. Prices provide information about pleasures, resources that are ready to use and the
opportunities of production while meeting this function. The second of the functions that the price undertakes is
directing people to places which ascribe the highest value to resources and to the production techniques with the
lowest cost. The third function of the price is about who will consume what and how much, that is the problem
of sharing the income (Tosun, 2002: 81-82).
International Journal of Economics and Financial Issues, Vol. 2, No. 3, 2012, pp.281-295
286
carry out more public expenditures as they cannot be used effectively and pursuant) and finally
governments appeal more monetizing and all of these can result in inflation (A-Marhubi, 2000: 1999;
Çelen, 2007: 112). Moreover, corruption can bring about a raise in general level of prices as an
additional cost element (Ekpo, 1985: 315).
As a consequence, it is possible to say that there is a strong relationship between inflation and
corruption (Al-Marhubi, 2000: 1999). This can be a two-way relationship, from inflation to corruption
and from corruption to inflation. The common belief in the relationship from inflation to corruption is
that the increase and rapid change in inflation results in escalation in corruptions.
3. Econometrical Analysis
3.1. Methodology
In this study, it was aimed to investigate the relationship between corruption and inflation, the
rule of law, the efficacy of the formation and implementation of government policies, the relationships
between the political stability of the governments and corruption through balanced panel data method.
The panel data method which has lots of advantages is preferred more and more in both macro and
micro level econometric studies (Davidson and MacKinnon, 1999: 296).
Having both cross sectional and time dimensions in panel data set provides some advantages
such as controlling the heterogeneity of the predictions, increasing the degree of freedom level and
reaching more reliable parameters (Baltagi, 2005: 4-9; Hsiao, 2006: 7). Panel data regression model is
shown below in the simplest way (Greene, 2003: 285):
(1)
expresses K (amount) variables that do not include fixed term. In term which shows
heterogeneity and individual effects, expresses the observable effects such as race, gender and place
or non-observable individual or group specific effects. This model general display of which is
presented can be enlarged depending on assumptions made about fixed term, slope coefficient and
error term. In the fixed effects model that is used in predicting the equation 1, it is assumed that each
section has unobservable and invariant characteristics over time and these characteristics are
considered by allowing each section to have different fixed term through dummy variables. In the
random effects model, however; it is accepted that the differences between the sections are accidental
and each section is allowed to have different fixed terms (Greene, 2003: 293). Panel least squares
model (LSV) is a very limited model as it does not take effects that are private for each section into
consideration. On the other hand, if you are certain that the neglected fixed effects and random effects
are independent, using the panel least squares method will provide more accurate results. For this
reason, the homogeneity of the effects belonging to the countries will be tested through Lagrange
Multipliers (LM) test which is suggested by Breusch and Pagan (1980) and (Brooks, 2008). Under the
dearth hypothesis that shows that the variances of the unit effects are zero, LM test has 1 degree of
freedom chi-square distribution (Greene, 2003: 299).
At the end of LM test, two models can be talked about if the dearth hypothesis is refused. These
models are named as the fixed effects model when non-observable effects and explanatory variables
are related with each other and as random effects model when non-observable effects and explanatory
variables are not related with each other. In order to choose the most accurate method in the study, the
model is predicted with random effects method first and then it is determined if the error term in the
model is related with the independent variables through Hausman test.
Before choosing the model, however; stability should be tested primarily in the series especially
which have long term size. In panel data analysis which realizes time series and cross-section analysis
together, variables should be stable so as not to be the cause to false relationships between variables.
In addition to this, the theory of panel data analysis is established on micro panels with “ big (the
number of countries) and small (time series)” asymptotic value. The asymptotic features of
predictors that are obtained are evaluated according to the assumption of “ ” for a certain T
value. In our study, as the time interval covers a short interval like and the number of sections
cover countries, it is assumed that the predictors that will be obtained at the end of the
prediction provide the asymptotic features.
Inflation and Corruption Relationship: Evidence from Panel Data in Developed and
Developing Countries
287
3.2. Model and Data Set
In this chapter of the study, the factors which are effective on corruption in countries were
investigated as in three different group levels according to their income levels through panel data
method. This discrimination that is made by the World Bank, the first group consists of 28 high-
income countries with over $ 12275 income; the second groups consists of 30 middle-income
countries with between $ 3975 and $ 12275 income and the third group consists of 39 low-income
countries with below $ 1005 income. As analyzing the years before 2002 results in the decrease in the
amount of cross sections, the period between 2002 and 2010 was selected as the research period and a
balanced panel was established with reference to annual data of this period. The data set which is used
in model analysis was obtained from the database of the World Development Indicators (WDI). In the
analysis, the panel regression model that will be used based on Braun and Tella (2004) and Al-
Marhubi (2000) is as follows:
(2)
In the regression equation above, the subscript of expresses the country, the subscript of
expresses the corruption rate
6
within the time period for the country at time.
indicates the inflation rate with the consumer prices, indicates the annual growth rate in Gross
Domestic Product. The variable of consists of exogenous variables of
which are thought to be related with the control of corruption in the countries. Summary information
and descriptive statistics about the variables are shown in Table 2.
Table 2. The Variables
Abbreviation Variable Explanation
CORRUP
The Index of Corruption
Control
The index value that shows the public power against
corruption
INFCON
Inflation with consumer prices
(annual %)
The annual inflation rate with consumer prices
GDPGRO
The growth in Gross Domestic
Product (annual %)
It shows the annual growth in Gross Domestic Product.
POLSTA
The Index of Political Stability
It is the index in which the political stability of governments
is measured.
REG
The Index of Legislation
Quality
The index about the stability and quality of the rules,
guidelines and regulations that the governments put into
action
RULE
The Index of the Rule of Law
The index values about the society’s keeping the rules, the
frequency of crime and violence, the quality of police and
court services
ACC
The Index of Responsibility
The index about expression, the right of choice, freedom of
association and accountability
Source: World Development Indicators (WDI), 2011. http://data.worldbank.org/indicator
4. Results
Three models based on assumptions about how the fixed term is are used so as to predict the
relationship between the variables. These are “pooled regression” (pooled OLS), “fixed effects” and
“random effects”, respectively. The first phase in choosing the correct method is carrying out the LM
test which tests the homogeneity of the country effects. The null hypothesis in which random effect
model turns into pooled regression model is tested if the variance of the unit effects is found as zero
through LM test.
H0: Pooled Regression, σ
2
α
= 0
H
1
: Random Effect, σ
2
α
> 0
It was concluded that the models cannot be predicted through pooled regression as the statistics
of LM > was significant at 1 % level by rejecting the hypothesis of (the statistics of tests
related with the models are shown in Appendix 3). Upon this result, the model is predicted through
random effects method first, and then it is tested via Hausman test to find out whether the error term in
the model is related with the independent variables in order to use the most accurate method (fixed and
random effects) in prediction. In all of the predictions, the assumption which claims that supposed
International Journal of Economics and Financial Issues, Vol. 2, No. 3, 2012, pp.281-295
288
error terms are not related with independent variables is rejected by Hausman test and the hypothesis
that presents that fixed affects are invalid altogether is also rejected in F tests. According to the results
of these two tests, fixed effects model provides the most reliable predictions. Table 3 introduces the
prediction results that are made by using this approach.
Table 3. Panel Data Prediction Results aimed at the factors that affect Corruption (PCSE Model)
Dependent
Variable
The Index of Corruption Control
Independent
Variables
Model I
(Underdeveloped
Countries)
Model II
(Developing Countries)
Model III
(Developed
Countries)
INFCON -0.0012[-1.64]** -0.0035[-1.74]** -0.0075[-1.31]**
GDPGRO -0.0038[-2.86]* 0.0024[1.18] 0.0101[2.05]*
POLSTA 0.0196[0.85] 0.0330[2.67]* -0.0028[-0.04]
REGUL 0.3675[3.92]* 0.3673[5.10]* 0.0535[1.87]**
RULE 0.3834[7.91]* 0.4380[4.66]* 0.4713[3.72]
ACC 0.2006[4.69]* 0.3582[3.17]* 0.2536[1.98]*
OPENNESS -0.0005[-0.48] 0.0008[0.93] 0.0013[1.64]**
Fixed Term -0.0756[-1.32] -0.2869[-2.93]* -0.2154[-0.99]
The number of
observations
351 270 224
The number of
countries
39 30 28
F Statistics 90.53 198.20 174.59
P value (F statistics) 0.0000* 0.0000* 0.0000*
2
R
0.920 0.956 0.975
The values in brackets are t statistics.
Statistically significant at * 5 % and ** 10 % significance levels.
In the next phase of the analysis, it was investigated if these three models have problems of
changing variance and autocorrelation. Woolridge autocorrelation test shows that the null hypothesis
which assumes that there is not a first-order autocorrelation in none of the models is rejected.
Regression coefficients that are predicted in case of changing variance and autocorrelation are
consistent but not effective. Two types of approaches are generally used in the literature in order to get
rid of these problems and obtain more reliable results. The first of them is the “Feasible Generalized
Least Squares (FGLS)” and the other one is “Panel Corrected Standard Errors (PCSE)” also known as
Prais-Winsten approach
2
. The study of Beck and Katz (1995) presented that PCSE approach provided
more reliable results in data sets in which the size of cross-section is bigger than the size of time (For
details, see Beck and Katz, 1995, 1996; and Okuyan and Tascı, 2010). This procedure also prevents
the loss of observation like in other methods by allowing the usage of first observation in every panel
(Gujarati, 1995). The prediction results that are obtained through PCSE method are consistent and the
problem of changing variance and autocorrelation also disappears (Tavares, 2001: 30). Since the size
of the cross-section (97 countries) is bigger than the size of time (9 years) in the data set of this study,
the predictions were made through PCSE method. In the PCSE approach, the deferred value of
dependent variable was added to the model so as to purge the model from first-order autocorrelation.
When the prediction results that are obtained from the panel data analysis are studied, it is seen
that there is a positive and statistically significant relationship between inflation (INFCON) and
corruption in all three groups of countries that are investigated in the scope of the analysis
3
.
2
For the studies which use FGLS and PCSE methods, see (Tavares, 2001: 30; HeeMin Kim et al., 2006: 38;
Rudra, 2005: 713; Hunter and Wu, 2010: 9; Kamps, 2006: 25).
3
One point that must not be ignored here while making interpretation is that as the control index of corruption
increases, corruptions in that country decreases and corruptions in that country increases as the control index of
corruption decreases according to the description of the index of corruption by WDI as explained above.
Inflation and Corruption Relationship: Evidence from Panel Data in Developed and
Developing Countries
289
Accordingly, an increase in the rate of inflation brings an increase in the rate of corruption
together in these three groups of countries which are investigated in the scope of this analysis. Briefly,
this obtained result proves the judgment which is very common about the relationship between
corruption and inflation in the literature
4
and which says: “high inflation is a factor that affects the
emergence of the income loss of individuals and groups, the decay in the income distribution, the
increase in the number of rent-seeking activities and the ambiguity”.
Furthermore, following relationships between the explanatory variables and the dependent
variable in the model were found:
The direction of the relationship between Domestic Income (GDPGRO) and corruption varies in
low and high-income countries and the increase in domestic income affects corruption in the positive
direction in low-income countries while this effect changes into the negative direction in high income
countries
5
.
Political Stability (POLSTA) which is a bigger problem in low and middle-income countries
than in high-income countries also showed its impact on corruption and the prediction results
introduced that the effect of political stability on corruption is important. The prediction results
obtained from the analysis presented that corruption is lower in high-income countries in which
political stability is relatively higher and this result was found as compatible with the one in the
literature.
The prediction results that are made for three groups of countries revealed that there is a
significant and negative relationship between the quality of regulations (REGUL) and corruption. The
value of the coefficient is much higher in low-income countries than in the other groups of countries.
Therefore, arrangements for preventing corruption behaviours in these countries provide more
effective results than in other countries.
The relationship between the Rule of Law (RULE) and corruption was found as low and
statistically significant in low-income countries. Active law systems which will ensure accountability
and transparency have the power to control the distortions that are possible in the execution system
The index of responsibility (ACC) about the accountability, freedom of association, freedom of
speech and the right of choice point to a statistically significant relationship in all three groups of
countries. While this effect is higher in developed countries which have relatively higher per capita
income level than other countries, it is lower in underdeveloped and developing countries. The results
show that the decrease in corruption will be higher in developed countries in which transparency,
accountability and freedoms are relatively higher.
5. Conclusion
Corruption which is described as the deviation from the law or ethical values for personal
interests is a fact that has many effects on the economic life in the scope of cause-result relationship.
The first negative thing that we can face with during the evaluation in the context of economic costs of
corruption is the decrease in investments by creating a negative effect on the investors and the
retardation in economic growth and development as a natural result of this. In addition, it is
emphasized in many studies in the literature that corruption affects the distribution and the effective
usage of the existing resources in the economy negatively and causes inflation and inequality in
income distribution. On the other hand, many economic and social factors are considered as the
Therefore, from this point on, the opposite of the coefficient signs is considered during the interpretations about
the control index of corruption (CORRUP) that is the dependent variable.
4
In the literature, it is expressed that corruption is high in countries in which inflation is high. See Mumcu
(1985); Al-Marhubi (2000); Abed and Davoodi (2002); Bahmani-Oskooee and Nasir (2002); Piplica (2011);
Getz ve Volkema (2001); Paldam (2002); Braun ve Di Tella (2004).
5
Bardhan (1997: 1327) claims that complex institutional structure cannot be managed effectively in the first
phases in which economic growth is seen in low-income countries and public servants have to undertake more
initiatives and this increases corruption. Besides, he thinks that corruption which raises in parallel with national
income in these countries is related with the defects in the tax system of the countries and believes that the
portion of high taxes coming from taxable income is high. Mauro (1997:85), however; cumbersome bureaucratic
regulations and high cost of doing business are effective in the positive relationship between growth and
corruption in low-income countries and entrepreneurs prefer illegal ways to accelerate the bureaucratic process.
International Journal of Economics and Financial Issues, Vol. 2, No. 3, 2012, pp.281-295
290
reasons of corruption. Accordingly, macro-economic factors such as low-wage and employment,
poverty, inequality in income distribution, inflation, lack of competitiveness of the economy,
insufficient economic growth can provide suitable opportunities for corruption to appear and spread.
Inflation which is described as continuous increase in the general level of prices and which is one of
the basic macro-economic performances is an important term which we face in the context of the
elements that cause corruption and the effects of corruption. It is underlined in the literature about the
economic analysis of corruption that these two terms are strongly connected with each other.
However, there are many various findings about the direction of this relationship. In this framework,
inflation is described as both the reason of corruption and a case that is caused by corruption.
Because of its characteristics such as reducing the level of real wages and minimizing the
purchasing power of money, inflation might entail the income loss of individuals and groups and
distortion of income distribution. These people who experience the income loss can appeal to different
methods to generate revenue so as to sustain their economic life conditions. In this direction, inflation
might cause an increase in corruption acts such as bribery, deceptions, jugglery, lobbying and rent-
seeking activities. Besides, continuous and sudden raises in the general level of prices might also
result in the increase in the ambiguity in economic life. The ambiguities in economic process are the
most important factors in the appearance and spread of corruption acts.
Conversely, a reduction in public revenues comes into discussion in economies in which
corruption is experienced intensively and this guides the governments to use items of income such as
coining money often. The negative situation that is caused by coining money becomes the experience
of living an inflationary process. Furthermore, coining money will be resorted again for the necessary
public incomes as the public incomes are not used effectively in economies in which corruption acts
such as lobbying and rent-seeking activities are seen commonly. Moreover, bribe payments can cause
an increase in the general level of prices as an additional cost factor in economies in which corruption
is seen.
The results of analysis showed that inflation increased the rent-seeking activities and corruption
in the countries as expected. When the effects of the other variables on corruption are studied, policies
aimed at developing the basic structure such as economic performance, political stability and legal
regulations in the countries will be really effective in preventing corruption.
In the context of cause and result, inflation and corruption are two concepts that are in
interaction between each other. In this particular study, the accuracy of this relationship in which
inflation is believed to be the reason of corruption, that is, from inflation to corruption, is tested in the
economies of 97 countries from different income groups about the period of 2002-2010 through panel
data method. As a result of the findings obtained in the study, it was concluded that there is a positive
and statistically significant relationship between inflation and corruption in the economies of all 97
countries, twenty eight from high-income level, thirty from middle-income level and thirty-nine from
low-income level. Accordingly, an increase in inflation causes an increase in corruption in the
countries of these three groups that is investigated in the scope of this analysis. This result verifies the
common view in the literature.
Consequently, it is necessary to apply effective and successful policies and methods in order
to remove the destruction that corruption, which is described as “the cancer of countries” by former
chief of World Bank, Wolfenson, made in societies (World Bank, 2000: 2). The most effective method
in struggling with corruption is to remove the reasons of corruption. It should be known that removing
the reasons of a problem is the primary solution method. From this perspective, a step which does not
aim at removing the reasons might result in new problems rather than providing solution methods. In
that case, the reasons of corruptions must be understood well first so as to define the strategies to fight
corruption. The main purpose of this study is answering the questions about the factors that cause
corruption in terms of “inflation-corruption” instead of putting forward a holistic perspective.
In this context, the case of inflation is one of the most important economic elements that need
to be considered in the struggle against corruption as also this particular study showed. Fighting with
inflation plays an effective role in improving both macro-economic performance and social life and it
becomes an important process in solving the problem of corruption economic and social costs of
which is accepted by everybody.
Inflation and Corruption Relationship: Evidence from Panel Data in Developed and
Developing Countries
291
References
Abed, G.T., Davoodi, H.R. (2002), Corruption, Structural Reforms, and Economic Performance in the
Transition Economies. G. T. Abed and S. Gupta (Ed.), Governance, Corruption, & Economic
Performance, International Monetary Fund, Publication Services, Washington, D.C.
Abed, G.T., Gupta, S. (2002), The Economics of Corruption: An Overview. G.T.Abed and S. Gupta
(Ed.) Governance, Corruption & Economic Performance, IMF Press, 119.
September,http://www.imf.org/external/pubs/ntf/2002/govern/index.htm
Ackerman, S.R. (1999), Corruption and Government: Causes, Consequences and Reform. Cambridge
University Press, Cambridge.
Adaman, F., Çarkoğlu, A., Şenatalar, B. (2001), Causes of Corruption in Turkey Through The Eyes of
Household and Recommendations on the Prevention of Corruption. TESEV Publisher, No. 24,
Istanbul
Aidt, T.S. (2003), Economic Analysis of Corruption: A Survey. The Economic Journal, 113, 632-652.
Akcay, S. (2001), Economic Analysis of Corruption in Developing Countries. Unpublished PhD.
Thesis, Afyon Kocatepe University, Instıtute of Socıal Scıences, Afyon.
Al-Marhubi, F.A. (2000), Corruption and Inflation. Economics Letters, 66, 199–202.
Andving, J.C., Fjeldstad, O.H., Sissener, T. (2000), Research on Corruption: A Policy Oriented
Survey. Commissioned by NORAD, Final Report, December, Oslo
Ata, A.Y., (2009), Opportunity and Motivation of Corruption in the Framework of Institutional
Economics: An Analysis on EU Countries. Economic Research Foundation Publications,
Istanbul.
Bahmani-Oskooee, M., Nasir, A. (2002), Corruption, Law and Order, Bureaucracy and Real
Exchange Rate. Economic Development and Cultural Change.
Baltagi, B.H. (2008), Econometric Analysis of Panel Data, Fourth Edition, London, John Wiley &
Sons Ltd.
Baltagi, B.H. (2005), Econometric Analysis of Panel Data. Third Edition, John Wiley & Sons Ltd.,
West Sussex, England.
Bardhan, P. (1997), Corruption and Development. Journal of Economic Literatüre, 35, 1320-1346.
Beck, N., Katz, J.N., (1995), What To Do (and Not To Do) With Time-Series Cross-Section Data.
American Political Science Review, 89(3), 634-647.
Beck, N., Katz, J.N. (1996), Nuisance vs. Substance: Specifying and Estimating Time-Series Cross-
Section Data. Political Analysis, 6(1), 1-36.
Braun, M., Di Tella, R. (2004), Inflation, Inflation Variability, and Corruption, Economics & Politics.
16(1), 77-100.
Breusch, T.S., Pagan, A.R. (1980), The Lagrange Multiplier Test and Its Applications to Model
Specification in Econometrics. Review of Economic Studies, 47, 239-253.
Brooks, C. (2008), Introductory Econometrics for Finance. Cambridge University Press, New York.
Celen, M. (2007), Economics of Corruption. Istanbul Chamber of Certified Public Accountants
Publications, No:24, Istanbul.
Ekpo, A.H. (1985), Corruption and Prices: A Theoretical Note. Nigerian Journal of Social Studies,
27(3), 305-317.
Getz, K.A., Volkema, R.J. (2001), Culture, Perceived Corruption and Economics: Model of
Predictors and Outcomes. Business and Society, 40(1), 7-30.
Gray, C.W., Kaufmann, D. (1998), Corruption and Development. Finance Development, 35(1) March,
7-10.
Gujarati, D. N. (1995), Basic Econometrics. 3rd ed. New York, NY: McGraw-Hill.
Gupta, S., Davoodi, H., Alonso-Terme, R. (1998), Does Corruption Affect Income Inequality and
Poverty, IMF Working Paper, WP/98/76.
Haider, A., Ud Din, M., Ghani, E. (2011), Consequences of Political Instability, Governance and
Bureaucratic Corruption on Inflation and Growth: The Case of Pakistan. MPRA Paper Series,
No. 35584.
Huang, H., Wei, S-J. (2003), Monetary Policies for Developing Countries: The Role of Corruption.
IMF Working Paper, WP/03/183.
International Journal of Economics and Financial Issues, Vol. 2, No. 3, 2012, pp.281-295
292
Hunter, J., Wu, F. (2010), Multifactor Consumption Based Asset Pricing Models Using the US Stock
Market as a Reference: Evidence from a Panel of Developed Economies. Economics and
Finance Working Paper Series, Working Paper No. 1017.
Husted, B.W. (1999), Wealth, Culture and Corruption. Journal of International Business Studies,
30(2), 339-360.
Kamps, A. (2006), The Euro as Invoicing Currency In International Trade. Workıng Paper Series, No
665/August.
Kim, H, Powell, G.B., Fording, R.C. (2006), Electoral Systems, Party Systems, and Substantive
Representation: An Analysis of Distortion in Western Democracies. Midwest Political Science
Association, Chicago.
Klitgaard, R. (1988), Controlling Corruption. University of California Press, England.
Lambsdorff, J.G. (2007), The Institutional Economics of Corruption and Reform. Cambridge
University Press.
Luo, Y. (2004), An Organizational Perspectives of Corruption, Management and Organization
Review, 1(1), 119-154.
Mauro, P. (1997), The Effects of Corruption on Growth, Investment, and Government
Expenditure, in Corruption in the World Economy. (Ed): Kimberly A. Elliott, Washington
DC: Institute for International Economics, 83-107.
Okuyan, H.A., Tascı, H.M. (2010), Determinants of Capital Structure: An Application to Turkey's Top
1000 Industrial Company. Journal of BRSA Banking and Financial Markets, 4(1), 105-120.
Oweye, O., Ibrahim B. (1996), The Macroeconomic Analysis of The Effects of Corruption on
Economic Growth of Developing Economics. Rivista Internazionale di Science Economiche
Commerciali, 43(1), 191-211.
Paldam, M. (2002), The CrossCountry Pattern of Corruption: Economics, Culture and Seesaw
Dynamic. European Journal of Political Economy, 18(2), 215-240.
Piplica, D. (2011), Corruption and Inflation In Transition Eu Member Countries. Ekon. Mısao I
Praksa Dbk. God XX., BR. 2, 469-506.
Rahmani, T., Yousefi, H. (2009), Corruption, Monetary Policy and Inflation: A Cross Country
Examination. Unpublished manuscript.
Rose-Ackerman, S. (1999), Corruption and Government: Causes, Consequences and Reform.
Cambridge University Press.
Rudra, N. (2005), Globalization and the Strengthening of Democracy in the Developing World.
American Journal of Political Science, 49(4), October, 704–730.
Samimi, A. J., Abedini, M., Abdollahi, M. (2012), Corruption And Inflation Tax in Developing
Countries. Middle-East Journal of Scientific Research, 11(3), 391-395.
Smith-Hillman, A.V. (2007), Competition Policy, Inflation and Corruption: Evidence From African
Economies. Applied Economics Letters, 14(9), 653-656.
Tavares, A., Antonio, F. (2001), State Constraints and Local Environmental Programs: Solid Waste
Management Policy Instrument Choice. Midwest Political Science Association meeting in
Chicago, April 19-21.
Tosun, M. U. (2002), A Public Failure Product: Corruption. Cingi, S., Guran, C., and Tosun M. U. (Ed.),
Corruption and Efficient State, Ankara Chamber of Commerce Publication, Ankara.
You, J-S., Khagram, S. (2005), A Comparative Study of Ineguality and Corruption. American
Sociological Review, 70, 136-157.
World Bank (2000), Anti-Corruption Policies and Program. by Jeff Huther and Anwar Shah,
Operations Evaluation Department, Washington D.C.
Inflation and Corruption Relationship: Evidence from Panel Data in Developed and
Developing Countries
293
Appendix 1. Countries Examined in the Analysis
Low-Income Countries
(Below $ 1,005)
Middle-Income
Countries
(Between $ 3,975 and $
12,275)
High-Income
Countries
(Over $ 12,275 income)
1
Bangladesh Azerbaijan Germany
2 Benin Bosnia and Herzegovina United States
3 Cote D’Ivor Bulgaria Austuria
4 Chad Algeria Bahamas
5 Indonesia China Barbuda
6 Ethiopia Dominican Republic Belgium
7 Gambia Equator Denmark
8 Ghana El Salvador Finland
9 Haiti Armenia France
10 India Morocco Netherlands
11 Cambodia Philippines Hong Kong
12 Cameroon Guatemala United Kingdom
13 Kenya South Africa Ireland
14 Kyrgyzstan Honduras Spain
15 Comoros Jamaica Israel
16 Congo Kazakhstan Sweden
17 Laos Colombia Swiss
18 Lesotho Macedonia Italy
19 Madagascar Egypt Iceland
20 Malawi Paraguay Japan
21 Mongolia Peru Canada
22 Moldova Romania Cyprus
23 Mauritania Russia Korea
24 Mozambique Serbia Luxembourg
25 Nepal Sri Lanka Macao
26 Nigeria Syria Norway
27 Nicaragua Thailand Singapore
28 Central African Republic Tonga Greece
29 Pakistan Tunusia
30 Rwanda Jordan
31 Senegal
32 Sudan
33 Tajikistan
34 Tanzania
35 Uganda
36 Ukraine
37 Vietnam
38 New Guinea
39 Zambia
International Journal of Economics and Financial Issues, Vol. 2, No. 3, 2012, pp.281-295
294
Appendix 2. Descriptive Statistics
Model I. Countries with Per Capita Income Under $1005
Average
Standard
Deviation a
Minimum Maximum
Number of
Observations
CORRUP
1.780477 0.398493 0.181179 4.483994 351
INFCON
8.506762 6.408785 -8.974740 44.39128 351
GDPGRO
5.206244 4.046238 -1.480000 33.62937 351
POLSTA
-0.775218 0.853653 -2.704945 1.077283 351
REG
-0.655620 0.384124 -1.629678 0.298009 351
RULE
-0.784727 0.456542 -1.885425 0.204862 351
ACC
-0.636808 0.589924 -1.774103 0.499093 351
Model II. Countries with Per Capita Income Between $3976-$12275
Average
Standard
Deviation a
Minimum Maximum
Number of
Observations
CORRUP
3.239396 0.636476 1.680163 6.287126 270
INFCON
5.445863 6.307829 -2.407303 44.96412 270
GDPGRO
3.776700 5.404933 -1.795499 37.99873 270
POLSTA
0.461891 0.637530 -1.409702 1.539648 270
REG
0.420762 0.700499 -1.581180 1.466499 270
RULE
0.315880 0.684800 -1.643012 1.601308 270
ACC
0.485192 0.812803 -1.909686 1.475959 270
Model III. Countries with Per Capita Income Over $12275
Average
Standard
Deviation a
Minimum Maximum
Number of
Observations
CORRUP
5.551137 0.660917 3.120842 9.590772 252
INFCON
2.116507 1.845538 -4.479938 12.67819 252
GDPGRO
2.447432 4.206869 -1.125498 26.91328 252
POLSTA
0.784479 0.609185 -1.729574 1.662776 252
REG
1.394526 0.369588 0.478880 1.986506 252
RULE
1.438284 0.442828 0.279476 2.014196 252
ACC
1.144969 0.464893 -0.445205 1.825517 252
Appendix 3. F, LM ve Hausman Tests
Wald F-test 421.99*
Fixed effects
Breusch-Pagan 322.10*
Random effects
Model I
Hausman 12.17**
Fixed effects/Random effects
Wald F-test 367.55*
Fixed effects
Breusch-Pagan 177.04*
Random effects
Model II
Hausman 16.12*
Fixed effects/Random effects
Wald F-test 228.22*
Fixed effects
Breusch-Pagan 84.96*
Random effects
Model III
Hausman 23.44*
Fixed effects/Random effects
Inflation and Corruption Relationship: Evidence from Panel Data in Developed and
Developing Countries
295
Appendix 4. Cross-Correlation Tables
Model I. Countries with Per Capita Income Under $1005
CORRUP INFCON GDPGRO POLSTA REG RULE ACC
CORRUP
1
INFCON
-0.0318 1
GDPGRO
0.0226 0.0255 1
POLSTA
0.5514 -0.0455 0.0490 1
REG
0.6198 -0.0227 0.0657 0.4863 1
RULE
0.8013 -0.0127 0.1463 0.6304 0.7066 1
ACC
0.5227 -0.0137 -0.0375 0.4905 0.6025 0.6424 1
Model II. Countries with Per Capita Income Between $3976-$12275
CORRUP INFCON GDPGRO POLSTA REG RULE ACC
CORRUP
1
INFCON
-0.2926 1
GDPGRO
-0.2395 -0.0166 1
POLSTA
0.7083 -0.4699 -0.0685 1
REG
0.7697 -0.3903 -0.1919 0.6173 1
RULE
0.8938 -0.3732 -0.2210 0.7773 0.8650 1
ACC
0.7972 -0.2042 -0.2047 0.6757 0.7545 0.7389 1
Model III. Countries with Per Capita Income Over $12275
CORRUP INFCON GDPGRO POLSTA REG RULE ACC
CORRUP
1
INFCON
-0.1120 1
GDPGRO
-0.0982 0.0796 1
POLSTA
0.5778 -0.0649 0.0302 1
REG
0.7956 -0.1477 0.0283 0.4481 1
RULE
0.9265 -0.1170 -0.1110 0.5949 0.8080 1
ACC
0.4376 -0.0091 -0.3400 0.3122 0.3465 0.5560 1
