Artikel 2
Artikel 2
Artikel 2
INTRODUCTION
The author is from the Athens University of Economics and Business. She would like
to thank the two anonymous referees, the participants at the 2006 EIASM International
Conference on Accounting, Auditing and Management in Public Sector Reforms in Siena
and the participants at the 5th Annual Conference of the Hellenic Finance and Accounting
Association in Thessalonica for their helpful remarks on an earlier version of the paper.
Address for correspondence: Sandra Cohen, Lecturer of Accounting, Athens University of
Economics and Business, Department of Accounting and Finance, 76 Patission Street, Athens
104-34, Greece.
e-mail: scohen@aueb.gr
C
265
266
COHEN
267
LITERATURE REVIEW
268
COHEN
C 2008 The Author
C 2008 Blackwell Publishing Ltd
Journal compilation
269
270
COHEN
tax bases, income levels and public investment share on total expenditures. They
also found that inefficiency was related to high shares of grants in total municipal
expenditures and population density. Moreover, Afonso et al. (2006) found that
the expenditure efficiency of the public sector at a country level is related to
non-discretionary factors.
The comparison of ratios among public entities to establish benchmarks is also
a useful outcome of the ratio analysis exercise. Nevertheless, only a few variables
in public sector finance have an absolute benchmark (Berne, 1992). As Chaney
et al. (2002) point out, benchmarking requires an appropriate comparison group,
like one that has similarities in size, geographic location, demographics, revenues
sources or services provided. Thus, as comparisons among local governments are
difficult, differences in demographics and local conditions should be incorporated
in the analysis (Finkler, 2005).
Finally, the values of financial ratios per se are influenced by the accounting principles that govern the preparation of accrual accounting financial
statements. Literature offers several instances where issues that relate to the
recognition and valuation of assets and the calculation of depreciation expenses
have provoked considerable debate (see, for example, Carnegie and West, 2003;
Christiaens and de Wielemaker, 2003; Gillibrand and Hilton, 1998; Heald and
Georgiou, 1995; Hepworth, 2003; Jones and Puglisi, 1997; McCrae and Aiken,
2000; Pallot, 1997 and 2001; and Perrin, 1998). However, in our case the adopted
accounting principles do not influence the cross-sectional analysis, as the same
rules apply without exception to all municipalities (PD 315/99) and the discretion
that can be exercised is extremely limited.
METHODOLOGY
Study Setting
In Greece there are 901 municipalities that are grouped in 52 prefectures
(counties) and 13 regions. The number of municipalities is large in relation to the
size (132,270 km2 ) and the population of the country (approximately 11 million
inhabitants). Also, there is significant variability in terms of their characteristics
such as population, geographical size, dependency on central government, etc.
Municipalities in Greece are responsible for a limited range of functions
when compared to other countries. The main activities of municipalities are
restricted to the provision of basic community services such as the development
and maintenance of local parks, the local registry, street lightening and cleaning,
refuse collection, provision of recreational services, roads maintenance and
repairing and limited transportation and health care services. The Greek
municipalities do not provide schooling and fire services.
Municipalities have their own budgets and follow budgeting principles and
rules that are common to all those entities binding the national government
C 2008 The Author
C 2008 Blackwell Publishing Ltd
Journal compilation
271
budget. They are also subject to several control mechanisms that limit mainly
expenditure choices. According to the Municipal and Communal Code, local
government expenditures are distinguished into mandatory and discretionary.
The first category is unavoidable. Wages and salaries, rents and debt charges
fall within this category. On the other hand, they have discretionary powers
over a range of expenditures. In recent years, municipalities have been given
more autonomy in levying taxes and determining fees for services but they
have in parallel experienced a significant reduction of subsidies and grants from
central Government. Municipalities are subsidised by Central Government for
the conduction of investments and for operating expenses coverage. Except for
the grants that are given to municipalities for the service of specific purposes,
the main amount of subsidies distributed to them is decided on a yearly basis by
the Ministry of Finance and MIPAD on an ad hoc basis. The later allocation aims
at smoothing out geographical and financial inequalities (Ministry of Finance,
2007, p. 146).
The compulsory adoption of accrual accounting since 1 January, 2000, was
restricted to municipalities that satisfied certain criteria (i.e., more than 5,000
citizens or revenues more than approximately 1.5 million euros). All other
municipalities could adopt accrual accounting optionally. After a considerable
delay, the majority of municipalities complied with the Law. According to data
published by the Hellenic Agency for Local Development and Local Government
(EETAA) in October 2005, 64% of the municipalities obliged to adopt accrual
accounting had published financial statements for the year 2003 (EETAA,
2005).
Ratio Selection
The goal of this study is to assess whether factors that are exogenous to the
municipalities control affect the values of the accrual accounting financial ratios
that can be used for financial performance assessment as well as for cross
sectional comparisons in analogy to the private sector paradigm. In other words,
we try to analyse if there is a systematic persistent effect of macro-economic
factors on financial ratios that could induce cross-sectional bias when comparing
heterogeneous municipalities that presumably operate in the same sector.
We selected for the purpose of our analysis a set of nine commonly used
financial ratios. Ratio selection is not a trivial matter but a rather contentious
one because information overlaps individual ratios (Barnes, 1987). This selection
aimed at portraying the financial condition and financial performance of
municipalities. Financial condition varies across a complex continuum with
very few local governments falling at either extreme and thus it could be
considered as a composite of factors and not a simple one-dimensional measure of
performance. As a result, ratio analysis consists of the measurement of different
factors, and consequently, permits the assessment of municipalities strengths
and weaknesses rather than a single assessment. It also involves comparisons, but
C
272
COHEN
in order for comparisons to be insightful they should be made only among similar
entities. In any case, in order to use ratios for performance assessment purposes,
reference points should also be developed. Moreover, these reference points
instead of being uniform across all type of municipalities should be properly
adjusted to reflect their idiosyncratic characteristics.
The selection of the financial ratios is based on both private sector and
public sector literature (ICMA, 2003; Berne, 1992; Anthony and Young, 2003;
and Finkler, 2005). It also lies among the lines used by the Ministry of
Finance to monitor at an aggregate level and extract statistical information for
the performance of public sector entities in years 19942000 under the cash
accounting regime (Ministry of Finance, 2002, pp. 8493; and Venieris and
Cohen, 2004). Additionally, it aimed at achieving a balanced representation
of the four broad categories of financial ratios, namely profitability ratios,
liquidity ratios, capital structure ratios and performance ratios within the
analysis framework.
The ratio name abbreviations as well as the way they are calculated are
presented in Table 1. The information content of the selected ratios is discussed
hereon.
Profitability Ratios
We use three profitability ratios in our analysis: the Return on Equity ratio
(ROE), the Return on Assets ratio (ROA) and the Profit Margin Ratio (PR). Even
though profitability is not a primary goal in the public sector the existence of a
reasonable surplus is necessary in order for a municipality to have enough funds
to finance its long-term capital investments. Alternatively, municipalities could
avoid the need for surpluses by relying exclusively on long-term debt. However,
this course of action is not an adequate approach and would influence their
solvency. Thus, profitability ratios provide a reflection of the efficiency in the use
of resources and the ability of management to finance growth. Contrary to the
expectations in the private sector, a small positive value for profitability ratios
could be considered as a favourable outcome. Large negative values, especially
when persistent, are indications of significant unfavourable financial prospects.
Liquidity Ratios
Liquidity defined as current assets to current liabilities (Current Ratio CR)
is an indicator of the municipalitys ability to pay its short-term obligations. A
low ratio may result in cash-flow problems that would require greater use of
short term borrowing to cover expenses. The liquidity indicator is informative of
the local governments ability to sustain a strong financial position. A liquidity
ratio of less than one hints at a relative poor liquidity (ICMA, 2003). Anthony
and Young (2003) argue that a value of around two usually corresponds to an
appropriate level. Thus, very low and very high values of the current ratio are
indicative of financial operating problems.
C 2008 The Author
C 2008 Blackwell Publishing Ltd
Journal compilation
C
D/E
LA
AT
Debt/equity
Assets turnover
OR/OE
CR
Current ratio
PR
Profit margin
OR/TR
ROA
Return on assets
ROE
Return on equity
Ratio Name
Ratio Calculation
ROE =
Ratio Calculation
Table 1
(9)
(8)
(7)
(6)
(5)
(4)
(3)
(2)
(1)
273
274
COHEN
Performance Ratios
In this final category we included the Assets Turnover ratio (AT) and two
ratios that are usually encountered in cash accounting regimes that interrelate
revenues and expenses, i.e., the Operating Revenues to Total Revenues ratio
(OR/TR) and the Operating Revenues to Operating Expenses ratio (OR/OE).
The Assets Turnover ratio permits the assessment of the efficient use of assets
within a municipality. Even though high rates of this ratio usually convey
favourable indications, its value can be influenced by the postponement of
investments that would increase the value of the denominator. The other two
ratios are frequently used for performance assessment under the cash basis and
thus it is highly probable that their use would continue despite the accounting
basis change. The higher the value of the OR/TR, the greater the municipalitys
independence from the State Budget, whereas, a low value is indicative of the
opposite. The higher the value of the OR/OE ratio the more sound financial
position characterizes a municipality. Moreover, as the aforementioned two
values are calculated on the basis of accrued revenues and expenses they
constitute proper indicators of the actual financial resources attained and used
during a yearly period. On the contrary, when they were used under the basis of
cash accounting they suffered from the inherent accuracy limitations in resources
measurement of the cash method.
275
private sector company is usually assessed against the industry norms on the
basis of financial ratios analysis albeit not without limitations. However, this is
not appropriate for local governments. The financial attractiveness of an area
and its growth opportunities that may stem from its geographical location, its
vicinity to resources, its political significance, etc. are expected to affect both
the number of inhabitants and its wealth. The last two variables are in turn
expected to influence underlying financial measures such as revenues, expenses,
assets and liabilities and ultimately the values of financial ratios.
Thus, in the context of ratio analysis we try to quantify the degree to which
factors that are exogenous to the municipalities control, such as their wealth
and size, influence their financial condition and performance. We use three
factors as a proxy of wealth: the gross domestic product (GDP), the real estate
values (PRICE) and the tourist activity (TOURIST) of the municipalities. The
gross domestic product is used as a proxy of the personal income level in the
municipality. The real estate values provide an indication of the citizens capacity
to pay taxes as they mirror the property tax base. Taxes are the primary source
of municipality revenues. The inclusion of the tourist activity as an independent
variable reflects an idiosyncratic factor of the economic environment in Greece.
In several areas the tourist industry is the main locomotive of local economic
activity. This is true mainly for the numerous Greek islands and rural areas that
have a close proximity to the sea or constitute winter resorts.
The size of the municipalities is measured by two variables: the population
(POP) of the municipality and whether the capital of the prefecture (CAPITAL)
is located in the municipality. Population is the most commonly used proxy for a
local governments size. However, as prefecture capital is usually the larger city
of a prefecture and the place where public administration services are gathered,
this variable was also included as an additional size variable.
The five above-mentioned selected independent variables additionally satisfy
two criteria. The first one is that this type of information is retrievable by
secondary sources and it is periodically updated. The second is that these
variables are straightforward, objective and easy to understand by both oversight
agencies and assessed municipalities.
Influence of Wealth
The economic health of a municipality is to a large extent dependent on the
income level of its citizens. Personal income is an important measure of citizens
ability to pay taxes. A high level of personal income generally means higher
property, sales and business taxes. Thus, the affluence of a municipality is
expected to be positively related to its capacity to generate revenues. A local
authority with wealthy citizens is able to levy, at least theoretically, higher local
taxes and charge higher prices for the services rendered. This prospect is also
expected to be positively related to its asset turnover; the invested capital would,
ceteris paribus, generate more revenues than in a less affluent place. Moreover,
C
276
COHEN
Influence of Size
According to Berne (1992) many social, political, economic and fiscal features
of governments appear to be related to size usually measured in terms of
population. The number of citizens within a municipality undoubtedly influences
the size of the municipality, e.g., number of municipal officers and employees,
volume of services offered, etc. When the size of a municipality increases,
bureaucratic problems become more intense and the execution of operations
and the management of resources may become less efficient. Thus, a large
municipality may experience increased requirements for efficient management
in order to operate properly and meet the service needs of the population
sufficiently. On the other hand, larger municipalities usually have adequate
human and material resources in order to sustain sophisticated systems that are
designed to support efficient decision making in relation to procedure execution
and current assets and short-term liabilities management. Moreover, the size
of a municipality is expected to be positively related to its bargaining power
during negotiations with both government and financial institutions in relation
to subsidies and fund-raising respectively. Finally, larger municipalities with
more capital needs are expected to exhibit long-term debt exposure in order to
finance their investments.
Data Collection
The analysis is based on all published accrual accounting financial statements
of municipalities for the years 2002 to 2004 available until March 2006. The
number of available financial statements is not stable throughout the time period
of the analysis. This is due to several reasons. Firstly, not all municipalities
commenced, as they were obliged, the implementation of accrual accounting on
time (i.e., year 2000). Thus, several municipalities published financial statements
that refer to the year 2003 or even the year 2004 for the first time. Secondly,
some municipalities have only published inaugural financial statements and then
they disregarded their obligation.2 Thirdly, a significant delay between the end
of the accounting period and the publication of the financial statements has
been witnessed. This delay has amounted to, on several occasions even two
C 2008 The Author
C 2008 Blackwell Publishing Ltd
Journal compilation
277
years. The source of the financial statements related data was the Ministry
of Interior Public Administration and Decentralization (MIPAD) where copies
of the financial statements were available.
The sources of information in relation to the macro-economic factors are the
following:
GDP: The Greek National Statistical Office does not keep GDP statistics on
a municipality level. However, such information is available on a prefecture
level. Thus, we used the GDP of the prefecture to which the municipality
belongs as a proxy for municipalitys GDP. Provisional data was available
for years 20022004.
POP: The number of inhabitants per municipality was retrieved from the
Greek National Statistical Office and refers to census 2001.
PRICE: The regional price level of real estate was gathered from different
sources. The main source was the official list issued by the Ministry of
Finance containing the prefixed prices for real estate transactions for
taxation purposes. However, such information was not available for several
municipalities for which the Ministry of Finance has not yet developed such
benchmarks.
TOURIST: The classification of municipalities as tourist is defined by Law,
which characterizes specific areas as tourist zones that enjoy tax and other
privileges.
CAPITAL: The information about the location of the prefecture capital
(e.g., metropolitan municipalities) was retrieved from the MIPAD.
The Model
The following multivariate regression model was used in order to assess the
relation between each financial ratio and macro economic factors.
Financial ratio = a 0 + b 1 log GDP + b 2 log pop + b 3 log price
+ b 4 tourist + b 5 capital + e
where:
Financial ratio is one of the following ratios: Return on equity, Return
on assets, Profit margin, Current ratio, Debt/Equity, Long term liabilities/Total assets, Assets turnover, Operating revenues / Total revenues
and Operating revenues/Operating expenses as defined in Table 1;
Log GDP is the logarithm of the gross domestic product (GDP) of the
municipality;
C
278
COHEN
Log pop is the logarithm of the population (POP) of the municipality (census
2001);
Log price is the logarithm of the property values (PRICE) in the municipality;
Tourist is a dummy variable taking the value 1 when the municipality has
significant tourist development and 0 otherwise;
Capital is a dummy variable taking the value 1 when the municipality hosts
the capital of the prefecture and 0 otherwise.
As the distribution of GDP, POP and PRICE variables was highly skewed, the
logarithm transformation was used. The transformation resulted in obtaining
satisfactory levels of normality and symmetry in terms of distribution of these
variables.
RESULTS
Table 2 describes the number of municipalities per region and per accounting
period in the sample for which financial statements data were available. The 497
municipality years correspond to 277 physical municipalities.
The statistical analysis was conducted on a yearly basis. In order to avoid
extreme ratio values, we excluded from the analysis all ratios with values that
fell out of 2 std. deviations. Table 3 presents the descriptive statistics of ratios
after the elimination of outliers and the descriptive statistics of independent
variables for the years 20022004.
Table 4 presents the Pearson correlations among the independent variables
for the three-year period. The Pearson correlations per year are qualitatively
the same. All independent variables are statistically significantly correlated but
none of the pairwise correlation coefficients exceeds the conventional threshold
of 0.800 that could cause multicollinearity problems (Gujarati, 1995). Moreover,
the fact that the logGDP and the Tourist variables that are both wealth variables
are negatively correlated indicates that the characteristic of wealth may take
several forms and it is not univocally defined.
The results of the OLS regressions are presented in Table 5 where Panels A to
C refer to the results of the years 2002 to 2004 respectively. The regressions have
been tested for collinearity problems. The rule of thumb is that there is evidence
of multicollinearity problems if the variance inflation factor (VIF) of a variable
exceeds 10 (Gujarati, 1995). The values of VIF of all explanatory variables in all
regressions in Table 5 did not exceed this value.
The results in Table 5 reveal that profitability ratios seem to be influenced
by two macro-economic factors; population and real estate prices. A positive
statistically significant relation at 5% statistical significance level of real estate
prices to all profitability ratios in 2002 and ROE in 2003 is evident. Also, a
C 2008 The Author
C 2008 Blackwell Publishing Ltd
Journal compilation
279
Table 2
Number of Municipalities Year Observations per Fiscal Year and
Region
Regions
2002
2003
2004
Total Number of
Municipality Years
per Region
Total Number
of Municipalities
per Region that have
Published Financial
Statements at Least
for One Year
Kriti
Dytiki Makedonia
Sterea Ellada
Dytiki Ellada
Anat. Makedonia & Thraki
Notio Aigaio
Ipeiros
Thessalia
Kentriki Makedonia
Attiki
Peloponnisos
Ionia Nisia
Voreio Aigaio
Total
1
2
2
3
5
4
3
16
32
31
9
6
0
114
7
6
15
24
10
10
15
36
64
41
15
11
0
254
4
4
15
21
6
9
4
14
27
11
8
6
0
129
12
12
32
48
21
23
22
66
123
83
32
23
0
497
8
6
21
31
10
13
15
38
65
42
17
11
0
277
280
COHEN
Table 3
Descriptive Statistics for Financial Ratios and Macro-economic Factors
for Years 20022004
Panel A: 2002
ROAa
ROEa
RPa
CRa
D/Ea
ATa
LAa
OR/TRa
OR/OEa
Pop (in thousands)
GDP (in billion )
Real estate (in 00/m2)
Tourist (value = 1)
Capital (value = 1)
Panel B: 2003
ROA
ROE
RP
CR
D/E
AT
LA
OR/TR
OR/OE
Pop (in thousands)
GDP (in billion )
Real estate (in 00/m2)
Tourist (value = 1)
Capital (value = 1)
Panel C: 2004
ROA
ROE
RP
CR
D/E
AT
LA
OR/TR
OR/OE
Pop (in thousands)
GDP (in billion )
Mean
Std.Dev.
Min.
Max.
0.0049
0.0124
0.0658
2.5757
0.1501
0.1232
0.0421
0.5455
0.4754
34.1056
18.8448
8.9989
0.0261
0.0334
0.1500
3.3445
0.1691
0.0908
0.0450
0.1548
0.1649
78.9749
22.5364
7.1927
0.0689
0.0955
0.4964
0.1184
0.0018
0.0105
0.0000
0.1394
0.0994
3.7790
0.6130
2.2000
0.0718
0.0803
0.2460
22.7571
0.7862
0.3513
0.1737
0.8391
0.8222
745.5140
54.5050
44.0205
99
99
104
111
109
104
103
106
103
114
114
111
35
19
0.0047
0.0020
0.0037
3.5129
0.1156
0.1101
0.0336
0.5134
0.4709
20.8488
12.9915
7.3639
0.0262
0.0326
0.1672
4.8379
0.1498
0.0803
0.0413
0.1649
0.1754
54.7211
20.8681
6.2055
0.0689
0.0992
0.4567
0.1096
0.0000
0.0017
0.0000
0.0948
0.0631
0.7480
0.3340
2.2000
0.0818
0.0932
0.4404
46.1100
0.8708
0.3407
0.1864
0.9134
0.8876
745.5140
59.1690
44.0205
240
241
243
249
247
245
240
246
242
254
254
232
69
25
0.0023
0.0067
0.0303
4.5697
0.1080
0.1060
0.0329
0.4824
0.4163
12.6432
8.4655
0.0272
0.0351
0.1382
8.1905
0.1368
0.0789
0.0356
0.1773
0.1713
13.4712
17.6804
0.0787
0.1013
0.4461
0.1308
0.0001
0.0062
0.0000
0.0899
0.0600
1.1680
0.3502
0.0789
0.0978
0.3445
60.3799
0.8096
0.3490
0.1470
0.8638
0.7994
66.0170
64.7040
123
124
120
128
125
126
121
126
124
129
129
C 2008 The Author
C 2008 Blackwell Publishing Ltd
Journal compilation
281
Table 3 (Continued)
Mean
Std.Dev.
Min.
Max.
6.8826
4.8497
2.2000
33.7491
104
35
11
Notes:
a
Ratio definitions are presented in Table 1.
Pop is the population of the municipality (census 2001).
GDP is the gross domestic product of the prefecture as proxy for the GDP for the municipality.
Price is the property values in the municipality.
Tourist is a dummy variable taking the value 1 when the municipality shows significant tourist
development and 0 otherwise.
Capital is a dummy variable taking the value 1 when the municipality hosts the capital of the
prefecture and 0 otherwise.
Table 4
Pearson Correlations Among Independent Variables
Log pop
Log GDP
Log price
Tourist
Capital
Log pop
Log GDP
Log price
Tourist
Capital
1.000
0.418
0.627
0.227
0.480
1.000
0.547
0.186
0.129
1.000
0.270
0.297
1.000
0.466
1.000
Notes:
Log GDP is the logarithm of the gross domestic product of the prefecture as proxy for the GDP for
the municipality.
Log pop is the logarithm of the population of the municipality (census 2001).
Log price is the logarithm of the property values in the municipality.
Tourist is a dummy variable taking the value 1 when the municipality shows significant tourist
development and 0 otherwise.
Capital is a dummy variable taking the value 1 when the municipality hosts the capital of the
prefecture and 0 otherwise.
Panel A: 2002
0.021
a0
(0.044)
Log pop
0.003
(0.398)
Log GDP
0.001
(0.852)
Log price
0.029
(0.043)
Tourist
0.012
(0.099)
Capital
0.011
(0.276)
F
2.272
(p-value)
(0.054)
0.063
R2 adj.
N
96
0.033
(0.015)
0.003
(0.519)
0.007
(0.317)
0.040
(0.030)
0.011
(0.234)
0.013
(0.304)
1.922
(0.098)
0.046
96
0.129
(0.029)
0.034
(0.137)
0.009
(0.770)
0.172
(0.033)
0.018
(0.665)
0.006
(0.918)
1.698
(0.143)
0.034
101
3.999
(0.002)
0.984
(0.043)
0.668
(0.320)
1.750
(0.304)
0.450
(0.606)
1.785
(0.140)
2.339
(0.047)
0.059
108
0.062
(0.293)
0.044
(0.059)
0.115
(0.000)
0.166
(0.042)
0.088
(0.036)
0.094
(0.101)
5.645
(0.000)
0.181
106
AT
0.046
(0.156)
0.009
(0.451)
0.063
(0.000)
0.017
(0.699)
0.046
(0.044)
0.000
(0.999)
6.019
(0.000)
0.201
101
Financial ratio = a 0 + b 1 log pop + b 2 log GDP + b 3 log price + b 4 tourist + b 5 capital + e
ROA
ROE
RP
CR
D/E
Table 5
0.023
(0.172)
0.011
(0.098)
0.029
(0.002)
0.048
(0.038)
0.030
(0.013)
0.018
(0.273)
4.165
(0.002)
0.138
100
LA
0.418
(0.000)
0.042
(0.045)
0.072
(0.014)
0.199
(0.008)
0.027
(0.468)
0.062
(0.235)
6.612
(0.000)
0.216
103
OR/TR
0.332
(0.000)
0.022
(0.342)
0.075
(0.021)
0.132
(0.107)
0.074
(0.079)
0.053
(0.355)
5.416
(0.000)
0.182
100
OR/OE
282
COHEN
C 2008 The Author
C 2008 Blackwell Publishing Ltd
Journal compilation
C
F
(p-value)
R2 adj.
N
Capital
Tourist
Log price
Log GDP
Log pop
Panel B: 2003
a0
0.003
(0.677)
0.005
(0.050)
0.001
(0.724)
0.019
(0.051)
0.004
(0.414)
0.001
(0.904)
1.667
(0.144)
0.015
219
0.001
(0.925)
0.012
(0.000)
0.002
(0.707)
0.042
(0.001)
0.005
(0.368)
0.003
(0.781)
4.098
(0.001)
0.066
220
0.012
(0.773)
0.021
(0.231)
0.001
(0.964)
0.070
(0.271)
0.035
(0.247)
0.025
(0.606)
1.114
(0.354)
0.003
222
5.718
(0.000)
1.683
(0.000)
0.801
(0.256)
3.056
(0.083)
0.870
(0.301)
1.884
(0.170)
3.661
(0.003)
0.056
227
0.018
(0.581)
0.058
(0.000)
0.100
(0.000)
0.147
(0.003)
0.063
(0.008)
0.057
(0.137)
14.604
(0.000)
0.233
225
AT
0.048
(0.007)
0.016
(0.035)
0.048
(0.000)
0.019
(0.489)
0.048
(0.000)
0.033
(0.116)
11.727
(0.000)
0.195
223
Financial ratio = a 0 + b 1 log pop + b 2 log GDP + b 3 log price + b 4 tourist + b 5 capital + e
ROA
ROE
RP
CR
D/E
Table 5 (Continued)
0.008
(0.405)
0.016
(0.000)
0.020
(0.001)
0.036
(0.014)
0.015
(0.029)
0.014
(0.217)
9.207
(0.000)
0.159
218
LA
0.401
(0.000)
0.010
(0.527)
0.085
(0.000)
0.079
(0.162)
0.059
(0.029)
0.060
(0.177)
9.811
(0.000)
0.165
224
OR/TR
0.344
(0.000)
0.002
(0.892)
0.085
(0.001)
0.079
(0.202)
0.061
(0.038)
0.029
(0.550)
8.650
(0.000)
0.149
220
OR/OE
283
0.003
(0.768)
0.004
(0.408)
0.006
(0.363)
0.016
(0.332)
0.002
(0.752)
0.004
(0.745)
0.768
(0.575)
0.012
99
0.016
(0.223)
0.001
(0.919)
0.002
(0.844)
0.035
(0.101)
0.007
(0.466)
0.013
(0.409)
1.241
(0.296)
0.012
99
0.001
(0.988)
0.009
(0.712)
0.027
(0.444)
0.048
(0.582)
0.003
(0.944)
0.004
(0.956)
0.337
(0.889)
0.036
97
8.077
(0.009)
2.958
(0.038)
2.139
(0.273)
4.657
(0.332)
0.426
(0.838)
2.718
(0.452)
1.772
(0.126)
0.036
103
0.046
(0.310)
0.059
(0.005)
0.088
(0.003)
0.012
(0.865)
0.000
(0.989)
0.076
(0.155)
8.385
(0.000)
0.272
100
0.029
(0.273)
0.016
(0.191)
0.048
(0.006)
0.009
(0.833)
0.053
(0.004)
0.022
(0.482)
6.782
(0.000)
0.224
101
AT
0.001
(0.921)
0.011
(0.077)
0.003
(0.709)
0.006
(0.778)
0.009
(0.332)
0.016
(0.325)
1.964
(0.091)
0.048
97
LA
0.382
(0.000)
0.018
(0.529)
0.088
(0.031)
0.082
(0.407)
0.127
(0.004)
0.023
(0.753)
4.315
(0.001)
0.142
101
OR/TR
0.295
(0.000)
0.006
(0.839)
0.091
(0.020)
0.043
(0.650)
0.103
(0.014)
0.058
(0.419)
4.729
(0.001)
0.160
99
OR/OE
Notes:
Financial ratio is one of the following ratios: Return on equity, Return on assets, Profit margin, Current ratio, Debt/Equity, Long term liabilities/Total assets,
Assets turnover, Operating revenues/Total revenues and Operating revenues/Operating expenses. Ratio definitions are presented in Table 1.
Log GDP is the logarithm of the gross domestic product of the prefecture as proxy for the GDP for the municipality.
Log pop is the logarithm of the population of the municipality (census 2001).
Log price is the logarithm of the property values in the municipality.
Tourist is a dummy variable taking the value 1 when the municipality shows significant tourist development and 0 otherwise.
Capital is a dummy variable taking the value 1 when the municipality hosts the capital of the prefecture and 0 otherwise.
P values are shown in parentheses.
Significant coefficients at 1% or 5% significance level are presented in bold type.
F
(p-value)
R2 adj.
N
Capital
Tourist
Log price
Log GDP
Log pop
Panel C: 2004
a0
Financial ratio = a 0 + b 1 log pop + b 2 log GDP + b 3 log price + b 4 tourist + b 5 capital + e
ROA
ROE
RP
CR
D/E
Table 5(Continued)
284
COHEN
C 2008 The Author
C 2008 Blackwell Publishing Ltd
Journal compilation
285
and GDP (2002: p = 0.002 and 2003: p < 0.001) but negatively associated with
property values (2002: p = 0.038 and 2003: p = 0.014). It is also positively related
at the 5% statistical significance level in years 2002 and 2003 to the tourist
development of the municipality. From these findings it can be inferred that
larger and wealthier municipalities tend to borrow more money from financial
institutions compared to smaller and less affluent ones. Another explanation
could be that only the municipalities with the above-mentioned profile (i.e.,
large and wealthy) seek long-term financial support from financial institutions.
Our results are particularly important as far as the OR/TR and OR/OE ratios
are concerned. These two ratios proved to be positively related mainly with GDP
(OR/TR, 2002: p = 0.014, 2003: p < 0.001 and 2004: p = 0.031; OR/OE, 2002:
p = 0.021, 2003: p < 0.001 and 2004: p = 0.020) and property values (OR/TR,
2002: p = 0.008), albeit not during the whole period of analysis. Population is
negatively related to OR/TR only in 2002 at 5% significance level (p = 0.045).
Finally, the development of a municipality in relation to tourism is statistically
positively related at the 5% significance level with the OR/TR ratio (p = 0.029
in 2003 and p = 0.004 in 2004) and the OR/OE ratio for years 2003 and 2004
(p = 0.038 in 2003 and p = 0.014 in 2004). Thus, our data reveal that the
more affluent municipalities rely less on government subsidies and exhibit an
improved operating revenues/operating expenses relationship. These findings
are consistent with our expectations of wealth effect on performance.
The finding that financial ratios are influenced by macro-economic factors was further tested by splitting the initial sample into two groups for
each control variable per year. More specifically, the municipalities that had
published financial statements in any given year (20022004) were split into
two groups on the basis of the median value of the three continuous control
variables: large vs small municipalities (in terms of POP) and wealthier vs
less wealthier municipalities (in terms of GDP and PRICE respectively). The
median values of the independent variables that were used on a yearly basis
for this purpose are different throughout the three-year period. The trend in
the median values is consistent with the Greek accrual accounting adoption
experience where larger and wealthier municipalities were the pioneers in
publishing accrual accounting statements and the smaller and less affluent ones
followed.
As shown in Table 6, the statistically significant differences in the mean values
of the financial ratios under analysis reveal that there are several dissimilarities
among the Greek municipalities that compose a highly heterogeneous mosaic
in terms of wealth and size. More specifically, wealth measured both in terms
of GDP and PRICE proved to be a statistically significant differentiation factor
in the mean values of almost all measured ratios with the exception of the
profitability ratios. The conclusion that can be drawn from Table 6 is that
wealthier and larger municipalities perform better in operating terms; they
borrow more and at the same time they manage their short-term assets and
liabilities more efficiently than their less affluent and smaller counterparts.
C
ROA
ROE
RP
CR
D/E
AT
LA
OR/TR
OR/OE
Ratioa
Mean
0.0047
0.103
0.0772
3.390
0.1048
0.1004
0.0327
0.5047
0.4426
Mean
0.0049
0.0142
0.5646
1.857
0.1914
0.1436
0.0508
0.5792
0.5040
Group A
Group B
GDPmedian
GDP<median
GDP
GDP
Panel A: 2002
Median = 3.442 (in billion )
0.0002
0.0038
0.0207
1.532
0.0866
0.0432
0.0180
0.0745
0.613
Diff
0.096
0.565
0.485
0.022
0.006
0.013
0.042
0.013
0.059
Significance
(p-value)
Group B
Pop<median
Pop
0.0028
0.0092
0.0589
2.3512
0.1821
0.1455
0.0492
0.5724
0.5205
Mean
0.0069
0.0153
0.0732
2.796
0.1197
0.1001
0.0352
0.5175
0.4276
Mean
Group A
Popmedian
Pop
0.0041
0.0061
0.0142
0.4448
0.0623
0.0453
0.0140
0.0548
0.0929
Diff
0.432
0.364
0.631
0.486
0.057
0.010
0.114
0.070
0.004
Significance
(p-value)
Group B
Price<median
Price
0.0014
0.0057
0.0312
2.581
0.1611
0.1577
0.0449
0.6040
0.5388
Mean
0.0110
0.0190
0.1058
2.619
0.1444
0.0911
0.0401
0.4910
0.4220
Mean
Group A
Pricemedian
Price
Diff
0.0124
0.1329
0.0745
0.0379
0.0167
0.0665
0.0048
0.1130
0.1167
Differences in the Mean Values of Financial Ratios Among Groups for Years 20022004
Table 6
0.019
0.053
0.013
0.954
0.615
0.000
0.600
0.000
0.000
Significance
(p-value)
286
COHEN
C 2008 The Author
C 2008 Blackwell Publishing Ltd
Journal compilation
C
Mean
0.0060
0.0028
0.0165
2.7791
0.1470
0.1290
0.0399
0.5695
0.5259
Ratioa
ROA
ROE
RP
CR
D/E
AT
LA
OR/TR
OR/OE
0.0035
0.0013
0.0093
4.2525
0.0854
0.0918
0.0275
0.4583
0.4186
Mean
Group A
Group B
GDPmedian
GDP<median
GDP
GDP
Panel B: 2003
Median = 2.447 (in billion )
0.0025
0.0015
0.0259
1.4733
0.0615
0.0371
0.0124
0.1111
0.1072
Diff
0.453
0.711
0.227
0.016
0.001
0.000
0.020
0.000
0.000
Significance
(p-value)
Group B
Pop<median
Pop
0.0051
0.0003
0.0066
2.925
0.1541
0.1257
0.0461
0.5443
0.5057
Mean
0.0044
0.0044
0.0141
4.1150
0.0780
0.0944
0.0216
0.4831
0.4373
Mean
Group A
Popmedian
Pop
0.0006
0.0047
0.0207
1.1898
0.0761
0.0313
0.0244
0.0612
0.0684
Diff
Table 6 (Continued)
0.841
0.259
0.334
0.054
0.000
0.002
0.000
0.003
0.002
Significance
(p-value)
Group B
Price<median
Price
0.0055
0.0021
0.0065
2.9996
0.1462
0.1307
0.0426
0.5744
0.5309
Mean
0.0039
0.0017
0.0142
3.6537
0.0915
0.0926
0.0268
0.4604
0.4217
Mean
Group A
Pricemedian
Price
0.0015
0.0004
0.0207
0.6540
0.0547
0.0381
0.0157
0.1140
0.1092
Diff
0.657
0.919
0.360
0.208
0.008
0.000
0.006
0.000
0.000
Significance
(p-value)
287
0.0016
0.0019
0.0204
5.7291
0.0796
0.0927
0.0331
0.4575
0.3910
0.0061
0.0111
0.0399
3.4459
0.1360
0.1196
0.0326
0.5082
0.4424
ROA
ROE
RP
CR
D/E
AT
LA
OR/TR
OR/OE
0.0078
0.0092
0.0195
2.2832
0.0564
0.0269
0.0004
0.0506
0.0513
Diff
0.110
0.143
0.441
0.115
0.021
0.055
0.946
0.109
0.095
Significance
(p-value)
Group B
Pop<median
Pop
0.0000
0.0061
0.0280
2.978
0.1445
0.1208
0.0416
0.5013
0.4478
Mean
0.0046
0.0073
0.0328
6.2116
0.0721
0.0912
0.0245
0.4642
0.3838
Mean
Group A
Popmedian
Pop
0.0046
0.0011
0.0048
3.2332
0.0723
0.0295
0.0171
0.0371
0.0639
Diff
Notes:
a
Ratio definitions are presented in Table 1.
Significant mean differences at 1% or 5% significance level are presented in bold type.
GDP is the the gross domestic product of the prefecture as proxy for the GDP for the municipality.
Pop is the population of the municipality (census 2001).
Price is the property values in the municipality.
Mean
Mean
Ratioa
Group A
Group B
GDPmedian
GDP<median
GDP
GDP
Panel C: 2004
Median = 2.507 (in billion )
Table 6 (Continued)
0.348
0.853
0.851
0.028
0.003
0.035
0.008
0.239
0.038
Significance
(p-value)
Group B
Price<median
Price
0.0027
0.0090
0.0395
3.7920
0.1448
0.1411
0.0391
0.5614
0.4997
Mean
0.0007
0.0010
0.0218
4.8825
0.0785
0.0840
0.0283
0.4446
0.3753
Mean
Group A
Pricemedian
Price
0.0020
0.0079
0.0176
1.0905
0.0663
0.0571
0.0108
0.1167
0.1244
Diff
0.713
0.252
0.514
0.493
0.024
0.000
0.131
0.000
0.000
Significance
(p-value)
288
COHEN
C 2008 The Author
C 2008 Blackwell Publishing Ltd
Journal compilation
289
Table 7
Cluster Composition in Number of Municipalities per Year
Year 2002
Year 2003
Year 2004
59
49
3
111
165
64
3
232
76
17
11
104
Cluster 1
Cluster 2
Cluster 3
Total number of municipalities
Our last test towards assessing the influence of macro- economic factors on
financial ratios was cluster analysis. We conducted hierarchical cluster analysis
by using as grouping variables only the three continuous independent variables
(i.e. LogGDP, Logpop and Logprice). The clustering procedure showed that in
every period the observations could be classified into three clusters. The number
of municipalities per cluster group per year is presented in Table 7.
A thorough analysis of the composition of the second cluster per year
revealed that all municipalities that are classified in this group belong to
only two prefectures; the prefecture of Attiki (where Athens, the capital of
Greece is located) and the prefecture of Thessaloniki (where Thessalonica, the
second larger city of Greece is located). The prefecture of Attiki accounts for
approximately the 38.0% of the countrys GDP (based on 20022004 data) and
the 34.3% of the total population. The prefecture of Thessaloniki generates
approximately the 11% of the GDP in Greece and hosts almost 1 mil. inhabitants.
The first cluster includes municipalities from prefectures all over Greece for all
three periods. The t-test of the differences in the mean values of ratios analysed
revealed, with the exception of profitability ratios, that the municipalities of
the second cluster (e.g., those that belong to the prefectures of Attiki and
Thessaloniki) perform better that those of the first cluster. The t-test statistics
are presented in Table 8. Thus, the location of the municipality and the macroeconomic characteristics of this geographical location are significant parameters
that influence financial performance.
CONCLUSIONS
290
COHEN
Table 8
Differences in the Mean Values of Financial Ratios Between
Municipalities Clusters for Years 20022004
Cluster 1
Mean
Cluster 2
Mean
Panel A: 2002
ROA
0.0056
ROE
0.0113
RP
0.0749
CR
3.1316
D/E
0.1068
AT
0.0945
LA
0.0347
OR/TR
0.5012
OR/OE
0.4359
0.0043
0.0150
0.0608
1.8947
0.2126
0.1592
0.0545
0.5989
0.5254
0.0013
0.0037
0.0141
1.237
0.1058
0.0647
0.0198
0.0977
0.0895
0.821
0.610
0.652
0.046
0.003
0.001
0.043
0.001
0.005
0.0042
0.0019
0.0024
3.6388
0.0846
0.0946
0.0281
0.4828
0.4367
0.0063
0.0013
0.0056
2.5203
0.2066
0.1562
0.0530
0.5993
0.5680
0.0021
0.0006
0.0032
1.1185
0.1220
0.0616
0.0249
0.1165
0.1313
0.648
0.920
0.892
0.024
0.000
0.000
0.002
0.000
0.000
Panel C: 2004
ROA
0.0025
ROE
0.0055
RP
0.0316
CR
5.1502
D/E
0.0779
AT
0.0996
LA
0.0294
OR/TR
0.4874
OR/OE
0.4155
0.0056
0.0099
0.0453
1.9165
0.2594
0.1694
0.0454
0.5762
0.5199
0.0031
0.0044
0.0137
3.2337
0.1815
0.0698
0.0160
0.0888
0.1044
0.669
0.742
0.706
0.005
0.019
0.002
0.254
0.053
0.019
Ratioa
Panel B: 2003
ROA
ROE
RP
CR
D/E
AT
LA
OR/TR
OR/OE
Mean Difference
Significance (p-value)
Notes:
a
Ratio definitions are presented in Table 1.
Significant mean differences at 1% or 5% significance level are presented in bold type.
Cluster 1 includes municipalities that are located to prefectures all over Greece except for Attiki
and Thesalloniki prefectures.
Cluster 2 includes municipalities that are explicitly located in Attiki and Thesalloniki prefectures.
have not yet been defined. Our research provides useful insights in relation to
the information content of accrual accounting financial ratios for cross sectional
municipalities comparisons and the limitations of unconstrained benchmarking
in the public sector. Moreover, our conclusions can be used as a basis for
public economic policy decision-making. Thus, in contrast to the private sector
C 2008 The Author
C 2008 Blackwell Publishing Ltd
Journal compilation
291
292
COHEN
maximum values in any given ratio and a qualitative assessment whether a given
municipality reflects favourable or unfavourable performance against desirable
outcomes. Finally, the financial performance of a municipality could be assessed
on a scale basis from very poor to very sound placed on a continuum where
intervals would correspond to specific ratio values. Moreover, the scale could be
different for small versus large municipalities, wealthy versus less affluent ones,
etc.
Finally, an issue that could provide a future research prospect is the analysis
of the impact of the new system of performance measurement and assessment
on actual financial condition improvement. The proposed performance measurement and assessment system discussed in the paper shares the characteristics
of a top-down system that is primarily oriented to accountability and control. It
remains to be proven whether such a system could serve as a vehicle to performance enhancement. Performance measurement in the public sector has tended
to be developed to provide legitimacy within an institutional environment rather
than to inform organizational change and service improvement.
As an epilogue, we could conclude that our survey is aligned to the broad
literature framework of assessing the usefulness of the application of private
sector methodologies in the public sector as described by the NPM regime and
contributes to the limitations faced during this transplant. The research findings
provide supportive evidence to the hypothesis that the assessment of public sector
entities on the basis of financial indicators is more complicated than expected
on the basis of private sector experience.
NOTES
1 Municipal Financial ratio analysis (www.moodys.com, accessed 31/7/2006).
2 According to our data 128 municipalities out of the 254 municipalities that had supplied MIPAD
with financial statements for 2003, published financial statements for the first time in 2003. Also
17 municipalities out of 129 that had published financial statements for 2004 published financial
statements for the first time in 2004. Moreover, an analysis of the data revealed that 61 out
of the total 277 municipalities that are included in our sample published financial statements
only for 2003. It is worth mentioning that financial statements for the whole three-year period
were available only for 25 municipalities in the sample.
REFERENCES
Afonso, A. et al. (2006), Public Sector Efficiency, Evidence for New EU Member States and
Emerging Markets, European Central Bank Working Paper Series 581.
Anthony, R. (1978), Financial Accounting in Nonbusiness Organizations: An Exploratory Study of Conceptual
Issues, FASB Research Report (Stamford).
and D. Young (2003), Management Control in Nonprofit Organizations (7th ed., McGraw-Hill
Irwin).
Athanassopoulos, A. and K. Triantis (1998), Assessing Aggregate Cost Efficiency and the Related
Policy Implications for Greek Local Municipalities, INFOR, Vol. 36, No. 3, pp. 6683.
Balaguer-Coll, M. et al. (2002), Efficiency and Quality in Local Government Management. The
Case of Spanish Local Authorities, Universitat Autonoma de Barcelona, WP 2002/2.
Barnes, P. (1987), The Analysis and Use of Financial Ratios: A Review Article, Journal of Business
Finance & Accounting, Vol. 14, No. 4, pp. 44962.
C 2008 The Author
C 2008 Blackwell Publishing Ltd
Journal compilation
293
Berne, R. (1992), Research Report: The Relationships Between Financial Reporting and the Measurement of
Financial Condition, GASB (Norwalk).
Brorstrom, B. (1998), Accrual Accounting, Politics and Politicians, Financial Accountability &
Management, Vol. 14, No. 4, pp. 31933.
Brusca Alijarde, M.I. (1997), The Usefulness of Financial Reporting in Spanish Local Governments, Financial Accountability & Management, Vol. 13, No. 1, pp. 1734.
Brusca, I. and V. Montesinos (2006), Are Citizens Significant Users of Government Financial
Information?, Public Money and Management (September), pp. 2059.
Butterworth, P. et al. (1989), The Local Authority Annual Report in the UK: An Explanatory
Study of Accounting Communication and Democracy, Financial Accountability & Management,
Vol. 5, No. 2, pp. 7387.
Carnegie, G.D. and B. P. West (2003), How Well Does Accrual Accounting Fit the Public Sector?,
Australian Journal of Public Administration, Vol. 62, No. 2, pp. 8386.
Chan, J.L. (2003), Government Accounting: An Assessment of Theory, Purposes and Standards,
Public Money and Management, Vol. 23, No. 1, pp. 1320.
Chaney, B. et al. (2002), The New Governmental Financial Reporting Model, Journal of Government
Financial Management (Spring), pp. 2731.
Christiaens, J. (1999), Financial Accounting Reform in Flemish Municipalities: An Empirical
Investigation, Financial Accountability & Management, Vol. 15, No. 1, pp. 2140.
(2001), Converging New Public Management Reforms and Diverging Accounting Practices
in Flemish Local Governments, Financial Accountability & Management, Vol. 17, No. 2, pp. 153
70.
and E. de Wielemaker (2003), Financial Accounting Reform in Flemish Universities: An
Empirical Study of the Implementation, Financial Accountability & Management, Vol. 19, No. 2,
pp. 185204.
Cohen, S. et al. (2007), Assessing IT as a Key Success Factor for Accrual Accounting Implementation in Greek Municipalities, Financial Accountability & Management, Vol. 23, No. 1, pp. 91111.
Coy, D. et al. (1997), Recipients of Public Sector Annual Reports: Theory and an Empirical Study
Compared, The British Accounting Review, Vol. 29, No. 1, pp. 10327.
De Borger, B. and K. Kerstens (1996), Cost Efficiency of Belgian Local Governments: A
Comparative Analysis of FDH, DEA, and Econometric Approaches, Regional Science and Urban
Economics, Vol. 26, pp. 14570.
Finkler, S. (2005), Financial Management for Public Health and Not-for-profit Organizations (2nd ed.,
Pearson Education).
Fischer, M. et al. (2004), Measuring Operations: An Analysis of US Private Colleges and
Universities Financial Statements, Financial Accountability & Management, Vol. 20, No. 4,
pp. 12951.
Gillibrand, A. and B. Hilton (1998), Resource Accounting and Budgeting: Principles, Concepts
and Practice The MoD Case, Public Money and Management, Vol. 18, No. 2, pp. 2128.
Goldman, F. and E. Brashares (1991), Performance and Accountability: Budget Reform in New
Zealand, Public Budgeting and Finance, Vol. 11, No. 4, pp. 7585.
Gujarati, D.N. (1995), Basic Econometrics (3rd ed., McGraw-Hill).
Guthrie, J. (1998), Application of Accrual Accounting in the Australian Public Sector Rhetoric
or Reality?, Financial Accountability & Management, Vol. 14, No. 1, pp. 119.
Heald, D. and G. Georgiou (1995), Resource Accounting: Valuation, Consolidation and Accounting
Regulation, Public Administration, Vol. 73, No. 4, pp. 57179.
Hellenic Agency for Local Development and Local Government (EETAA) (2005) Double Entry
Accounting in Municipalities (EETAA, Athens) (in Greek).
Hepworth, N. (2003), Preconditions for Successful Implementation of Accrual Accounting in
Central Government, Public Money and Management, Vol. 23, No. 1, pp. 3743.
Hodges, R. and H. Mellet (2003), Reporting Public Sector Financial Results, Public Management
Review, Vol. 5, No. 1, pp. 99113.
Hoque, Z. and J. Moll (2001), Public Sector Reform: Implications for Accounting, Accountability
and Performance of State-Owned Entities an Australian Perspective, The International Journal
of Public Sector Management, Vol. 14, No. 4, pp. 30426.
International City/Country Management Association (2003), Evaluating Financial Condition: A
Handbook for Local Government (ICMA, Washington).
C
294
COHEN
C 2008 The Author
C 2008 Blackwell Publishing Ltd
Journal compilation