Transcript Slides - Croatian National Bank

Corporate Governance and
Effective Central Banking:
Cross Country Empirical Evidence
Iftekhar Hasan*
Loretta J. Mester**
* Rensselaer Polytechnic Institute
and Bank of Finland
** Federal Reserve Bank of Philadelphia
and University of Pennsylvania
Dubrovnik, June 28, 2007
1
Motivation
The legal and institutional framework governing
central banks around the world has undergone
significant changes in recent years

Implementation of monetary policy without
political influence, including institutional changes
to make monetary policy more effective and
transparent (e.g., Reserve Bank of New Zealand,
1989; Bank of England, 1997; Federal Reserve in
the U.S., 2006)
2
Motivation – cont’
There has also been discussion focusing on
central bank supervising effectiveness in the
context of large corporate bankruptcies with links
to the financial sector
These issues have increased the interest in better
understanding the role of good governance in
improving the effectiveness of central banks and
yielding improved economic outcomes (e.g., Lybek
& Morris 2007; Frissel, Roszbach & Spagnolo
2007)
3
Motivation – cont’
Recent empirical work suggests a positive
association between the quality of central bank
regulation and supervision and nations’ economic
growth, due to efficient access to and allocation of
financial resources (e.g., Laeven and Levine 2003;
Sapienza and Zingales 2003)
However, the design of the optimal contract, the
possible trade-off between achieving monetary policy
goals and financial stability objectives and the link
between central bank independence, accountability
and supervision performance have received only
marginal attention.
4
The purpose of this paper
Is there a significant statistical relationship
between central bank governance and
institutional structure and the economic
outcomes that reflect the performance of
central banks?
Does this relationship differ across countries
in different stages of economic development?
5
Our contribution
Add to the growing literature on
governance, organizational form, and
central bank performance


Determine whether measures of governance
and organizational form are associated with
better economic outcomes, while literature
has focused on developing governance
measures for central banks
Provide cross country evidence, while
literature has focused on developed countries
6
Review of literature
Central bank responsibilities and corporate governance – key
roles:
 Central banks’ key roles: macroeconomic stability (low
inflation), stable exchange rates, and the fostering of
maximum sustainable growth (See, e.g., Caprio & Dimitri
1995; Mayes 1997; Leybek 2002; Tuladhar 2005).
 Most central banks are responsible for stability of the
payment and settlement system (e.g., Healy 2001).
 Several banks have some responsibility for directly
supervising and examining commercial banks for safety and
soundness.
 Many central banks deliver banking services (cash, check,
credit etc. e.g., Flannery 1996) – given the technological
change in the payment services has led consolidation of
check processing sites within the U.S. Fed has led the Fed
to think its corporate governance.
 Frisell, Roszbach & Spagnolo (2007) – in a survey of mostly
European countries found most frequently cited central bank
roles were monetary policy, financial stability, payment
system and supervisory objectives.
7
Review of literature – cont’
Why is the governance of central banks more complex
than that of other corporations?



The multiplicity of central bank goals and the difficulty of
measuring performance across the central bank’s goals
suggest that transparency may not be easily achieved
(Frisell, Roszbach & Spagnolo 2007)
There is also intrinsic opaqueness of the banking
business that reduces the effectiveness of governance
mechanisms (Adams & Mehran 2003; Caprio and
Levine 2004)
In addition, as a monopoly provider of public goods (a
stable currency and a well-functioning payment system),
a central bank is insulated from the disciplining forces of
product market competition and the market for
corporate control (Frisell et al. 2007)
8
Review of literature – cont’
Are there ways to organize the central bank that
would lead to better incentives and economic
outcomes?


Numerous studies demonstrate a clear link between
political and monetary instability (e.g., Bagheri &
Habibi 1998)
Independence from the government might enhance
central bank effectiveness (e.g., Alesina & Summers
1993; Lybek 1999; Fischer 2005; Cukierman 2005)
9
Review of literature – cont’



Monetary expansions driven by political factors seem
to be the main cause of high inflation in transition
economies (Maliszewski 2000)
Severe fiscal imbalances and narrow financial markets
lead inevitably to monetary deficit financing (Bagheri &
Habibi 1998)
Delegating monetary authority to central banks that
are highly averse to inflation may reduce ‘inflationary
bias’ (Maliszewski 2000)
10
Review of literature – cont’


Therefore, institutional devices such as central bank
independence and strong governance could
potentially impose financial discipline on policy
makers and restrict them from short-sighted
monetary expansion (Lybek 1999)
Board size, tenure length and turnover of the board
chair, board independence – evidence exists for
corporations, but the impact on central bank
performance is not clear a priori and empirically
limited (e.g., Dreher, Sturm & de Hann 2006;
Cukierman, Webb and Neyapti 1992)
11
Data
Central banks’ websites
Information given upon request from some central banks
Thomson’s Bankscope database
IMF international financial statistics
Morgan Stanley-Central Bank Directory (several issues)
BIS publications of blue books, orange books, individual
annual reports
World development indicators
Annual data 1996-2000
Include countries whose central banks were established
in 1993 or earlier
Classify countries into 3 groups: transition, developing
and developed economies
12
Sample by countries
Transition Economies: Albania, Armenia, Belarus, China,
Croatia, Czech Republic, Estonia, Georgia, Hungary,
Kazakhstan, Latvia, Lithuania, Moldova, Mongolia, Poland,
Russia, Slovakia, Slovenia, Ukraine (19)
Developing Economies: Aruba, Bahamas, Bahrain, Barbados,
Belize, Botswana, Brazil, Chile, Columbia, Costa Rica,
Dominican Republic, Ecuador, Egypt, El Salvador, Ethopia, Fiji,
Guatemala, Haiti, India, Indonesia, Jordan, Kenya, Kuwait,
Lebanon, Lesotho, Macau, Malawi, Malta, Mexico, Morocco,
Mozambique, Nepal, Nicaragua, Nigeria, Oman, Pakistan, Peru,
Philippines, Saudi Arabia, Sierra Leone, South Africa, Sri Lanka,
Taiwan, Tanzania, Trinidad and Tobago, Turkey, Uganda, United
Arab Emirates, Uruguay, Zambia, Zimbabwe (51)
Developed Economies: Australia, Austria, Belgium, Canada,
Cyprus, Denmark, Finland, France, Germany, Greece, Hong
Kong, Iceland, Ireland, Israel, Italy, Japan, Korea, Netherlands,
New Zealand, Norway, Portugal, Singapore, Spain, Sweden,
Switzerland, United Kingdom, United States (27)
13
Variables
Performance measures (by monetary goal):
 Price stability goal
INFL – annual CPI inflation rate in country i, year t (based
on WDI 2005)
ABSINFL – absolute value of annual CPI inflation rate
(WDI)
STDDEVNFL – standard deviation of annual CPI inflation
rate over the previous 3 years
H_MONETARY - an index measuring the success of the
country’s monetary policy based on the weighted average
inflation over the most recent three years and the degree
to which a country imposes price controls, as determined
by the Heritage Foundation as part of its Index of
Economic Freedom (0-100 scale, higher score is better
performance)
14
Variables – cont’

Real output goal
RGDPGR – Real GDP growth (WDI)
STDDEVRGDPGR – Standard deviation of real
GDP growth over the previous 3 years (WDI)
WSDINFLGDPGR - an equally weighted average
of the standard deviation of annual CPI inflation
over the previous 3 years and the standard
deviation of annual real GDP growth over the
previous 3 years (WDI)
15
Variables – cont’

Financial stability goal
PROBLOAN - problem loan ratio = dollar volume
of problem loans as a percent of dollar volume of
total loans (Bankscope database)
VAREXRATE – standard deviation of exchange
rate (IFS Statistics)
16
Variables – cont’
Central bank characteristics:




Focus on measures related to governance structure
that could potentially affect the effectiveness of the
central bank in achieving its goals as reflected in our
performance measures
INDEPENDENT = 1 if the central bank has autonomy from
the government in implementing monetary policy, and 0
otherwise (Individual Websites, E-mails, IFS, and
Literature)
DIRECTORS – number of directors on the central bank’s
board in country i, time t (Morgan Stanley Central Bank
Directory, Individual Websites and E-mails)
DIRECTORS_DIRS – percentage of outside directors
on the central bank’s governing board, country i time t
17
Variables – cont’
Central bank characteristics – cont’:
 GOV_TURNOVER = Average rate of turnover of central
bank governors since 1993, measured as number of unserved years as percentage of term of the governor
divided by total number of governors since 1993
(Websites and E-mails)
 GOV_TERM_INDET = 1 if the length of the governor’s
term in country i in year t is undetermined and 0 otherwise
(Morgan Stanley Central Bank Directory, Individual
Websites and E-mails)
 GOV_TERM_DET = 0 if the governor’s term is
undetermined and = number of years in the governor’s
term in country i time t otherwise (Morgan Stanley Central
Bank Directory, Individual Websites and E-mails)
 SUP_MON = 1 if the central bank has bank supervisory
responsibilities as well as monetary responsibilities, and 0
otherwise
18
Variables – cont’
Central bank characteristics – cont’:

AGE – the number of years since the founding of the
central bank (Morgan Stanley Central Bank Directory,
Individual Central Bank Websites, and Direct
Correspondence via email with the Central Banks) (NOTE:
we do not use age in the regressions)
19
Results – descriptive statistics
MEANS
MEAN DIFFS
Trans
Dvlping
Dvlped
Trans
Vs.
Dvlping
Trans
Vs.
Dvlped
Dvlping
Vs.
Dvlped
INFL
20.01
10.33
2.19
9.69**
17.82***
8.14***
ABSINFL
20.08
10.52
2.33
9.56**
17.75***
8.19***
STDDEVINFL
78.42
9.68
0.88
68.75**
77.54**
8.8*
H_MONETARY
49.49
69.85
83.8
−20.36*** −34.32***
−13.95***
RGDPGR
3.77
3.56
3.59
0.21
0.18
−0.03
STDDEVRGDPGR
3.51
2.64
1.27
0.87**
2.24***
1.37***
WSDINFLGDPGR
40.97
6.28
1.05
34.69**
39.92**
5.23**
PROBLOAN
6.58
5.25
3.82
1.33**
2.76***
1.43***
VAR_EXRATE
7.10
34.51
5.13
−27.41*
1.97
29.38**
20
Results – descriptive statistics, cont’
MEAN
MEAN DIFF
Trans
Dvlping
Dvlped
Trans
Vs.
Dvlping
Trans
Vs.
Dvlped
Dvlping
Vs.
Dvlped
DIRECTORS
8.22
7.13
9.48
1.09***
−1.26**
−2.35***
OUTSIDEDIRS_DIRS
14.15
27.17
16.63
−13.02*** −2.48
10.54***
FEMALEDIRS_DIRS
12.69
8.07
10.02
−4.62***
−2.66
−1.96
GOV_TERM_INDET
0.052
0.096
0.111
−0.05
−0.06
−0.01
GOV_TERM_DET
5.58
3.94
5.07
2.03***
0.51**
−1.13***
GOV_TURNOVER
0.292
0.288
0.15
0.03
0.14***
0.14***
INDEPENDENT
0.211
0.077
0.777
0.13***
−0.57***
−0.70***
SUP_MON
0.421
0.31
0.41
0.11**
0.01
−0.10**
AGE
30.63
44.46
119.55
−13.83*** −88.92*** −75.09***
21
Difference-in-means analysis
Price stability goals: transition economies exhibit the
highest inflation and inflation variability levels,
whereas developed countries exhibit the lowest
levels.

Price stability goals were more successfully attained
in developed economies
Growth goals: growth levels are the highest in
transition economies and the lowest in developed
countries, but the differences are not significant;
developed countries have significantly more stable
growth.

Developed economies were more successful in
achieving price and output stabilization objectives
22
Difference-in-means analysis – cont’
Financial stability goals:


Transition economies exhibit the highest rate of
problematic loans while developed countries exhibit the
lowest rates, but the rates are fairly low in all regions.
Exchange rates were considerably less stable in the
developing countries compared with transition and
developed countries
23
Difference-in-means – cont’
Central bank characteristics:




Central banks in developed countries tend to be older,
with greater supervising responsibilities, greater
independence, and a larger number of directors
Central banks in developing countries seem to have
more outside directors
Central banks in transition economies seem to have
more female directors
Central banks in transition and developing economies
seem to have a higher governor turnover than those
in developed countries
24
Relationship between central bank characteristics
and performance differ across country groups
F-Tests
Trans =
Dvlping
Trans =
Dvlped
Dvlping =
Dvlped
Trans =
Dvlping=
Dvlped
INFL
9.01***
4.87***
2.38**
4.90***
STDDEVINFL
7.55***
6.66***
0.06
4.34***
ABSINFL
8.91***
4.80***
2.45**
4.87***
RGDPGR
2.84***
3.17***
0.90
1.96**
STDDEVRGDPGR
4.73***
4.14***
0.90
2.88***
WSDINFLGDPGR
6.93***
6.18***
0.06
4.08***
H_MONETARY
18.33***
12.73***
3.50***
10.55***
PROBLOAN
0.65
1.03
1.14
0.99
VAR_EXRATE
1.01
0.02
1.65
1.09
25
Results – Regression analysis
26
Price stability goal performance
In developed countries, a greater number of
directors is associated with an increase in
inflation

Consistent with a large board size hindering
decision and individual accountability
An undetermined (or long) governor term is
associated with lower inflation

Consistent with less political dependence and
government intervention that might produce
better inflation results
27
Price stability goal performance –cont’
Central bank involvement in both supervision
and monetary policy might lead to worse
inflation outcomes in transition and developed
economies.
Central bank independence appears to be
significantly associated with lower inflation in
transition countries and higher inflation in other
countries. We find no significant association
with inflation volatility.

Unexpected result, contrary to the received wisdom.
28
Results – regression cont’
29
Output goal performance
The better measure may be the variability of GDP
growth

Monetary policy cannot affect output level in the long
run
Bank independence is associated with greater
growth stability


This is opposite of what one might expect if there is a
short-run tradeoff between inflation and output
variability and the government favors stabilizing
output rather than inflation.
Our results suggest independent central banks do not
act in a way that neglects output stabilization
30
Results – regression cont’
31
Financial stability
There is little significant relationship between central
bank characteristics and the problem loan ratio – the
adjusted R-squards are very low.
For transition countries, the number of directors is
positively related to exchange rate variability, while the
share of outside directors is negatively related to this
measure

Combined with previous findings, these results suggest
that for transition countries, more directors may not be
helpful to central bank performance, but more outside
directors might be conducive.
32
Conclusions-1
We ask two simple questions:
 (1) Is there a significant relationship between
central bank governance and institutional
characteristics and the economic outcomes
that reflect performance of central banks?

(2) Do these relationships differ across central
banks in operating countries at different stages
of economic development?
33
Conclusions-2
(1) Is there a significant relationship between central bank
governance and institutional characteristics and the
economic outcomes that reflect performance of central
banks?
 We do find some significant associations, however, we
find no strong definitive conclusion that central bank
organizational structure has strong correlations with
economic performance, either positively or negatively.
(2) Do these relationships differ across central banks in
operating countries at different stages of economic
development?
 We find significant differences in the relationship
between performance and central bank governance
and organizational characteristics across countries at
different stages of economic development.
34
Conclusions-3
Central bank independence is not always
significantly related to performance and in some
cases in developing and developed countries the
relationship is the opposite of what one might
expect.




Independence is not significantly related to inflation
variability
In developed countries, independence is significantly
associated with lower output variation and with lower
weighted price-and-output variation, but with a higher
inflation level.
In developing countries, independence is significantly
associated with a higher inflation level.
In transition economies, independence is significantly
associated with lower output variation and a lower
inflation level.
35
Conclusions-4
The size of the board and the percentage of outside
directors on the board do not appear to have a strong
correlation with performance across our performance
measures.
Central bank involvement in both supervision and
monetary policy might lead to worse inflation outcomes in
transition and developed economies
36
Conclusions-5
Caution in interpretation:


Preliminary results – still collecting data
Relatively short time frame in our sample
So lack of strong significance could merely reflect the
lack of a long enough time frame over which there was
been enough variation in economic outcomes.
Or could provide an explanation of Lybek and Morris’s
(2004) finding that there is little consensus among
central banks regarding the structure, size, and
composition of their governing bodies.

Several of the associations we find are sufficiently
surprising as to merit further exploration.
37