Transcript Foreign Entry and the Mexican Banking System, 1997-2004

Measuring Bank Regulation and Supervision:
Lessons from Economic History
Stephen Haber
Stanford University
Presented at the World Bank, October 26, 2007
What is Agreed Upon:
There is a causal relationship between
financial development and economic
growth: King and Levine 1993; Levine 1997,
1998, 2005; Levine and Zervos 1998; Levine,
Loayza, and Beck 2000; Rajan and Zingales 1998;
Gerschenkron 1962; Rousseau and Sylla 2003.
What is not agreed upon:
How do we encourage financial development?.
1. Stronger supervision and regulation?
2. Reforms to basic institutions?
Might #1, in the absence of #2, actually make things
worse?
Barth, Caprio, and Levine are in the last camp
Two Stage Least Square Regressions
Entry
Applications Activity
Denied
Restrictions
Gov't
Owned
Banks
Executive Constraints
-0.122***
-0.444***
-0.064**
Controls for Legal Origin
Yes
Yes
Yes
Instruments included: Latitude, Year of Indpendence,
Initial Executive Competition, Initial Executive Openness,
Initial Executive Constraints, fraction of population Catholic,
fraction Muslim, fraction non-protestant.
Does the BCL view hold when we look
at variance within countries over time
Regressions of Private Credit/GDP on Political
Institutions, 1960-2005
Robust T Statistics in Parentheses
Constraints on
Executive
Constant
Observations
Number of countries
R-squared
Pooled OLS
0.048
[6.53]***
OLS with Country
Dummies
0.015
[5.39]***
0.073
[3.35]***
3428
131
0.18
3428
131
0.73
* significant at 10%; ** significant at 5%; *** significant at 1%
Driscoll-Kraay Standard Errors (Robust to AR1 serial correlation)
How can we know whether the BCL
regressions are detecting causal relationships?
One method is to use history as a natural laboratory,
to see if the hypothesized causal relationships
can be seen to operate in real world cases.
I am therefore going to focus on a single case—but
that that has had five separate experiments in the
creation of a banking system: Mexico, 18802007.
Mexico’s first experiment:
The Porfirio Diaz dictatorship, 1876-1911
Getting bankers to deploy their capital meant
A. centralizing supervision
B. granting special privileges
C. Limiting entry.
The results:
A. segmented monopolies.
B. rent sharing with politicians.
Bank Shareholders Earned Rents
Ratio of Market to Book Values (weighted average), Mexican Banks
2.50
2.00
1.50
1.00
0.50
0.00
1901
1902
1903
1904
1905
1906
1907
Source: Maurer and Haber 2007.
1908
1909
1910
1911
Mexico’s Banking System Grew
The Mexican Banking Industry, 1897-1913
35
Number of Banks
30
25
Assets as % GDP
20
15
10
Deposits as % Assets
5
-
97
18
98
18
99
18
00
19
01
19
02
19
19
03
04
19
05
19
19
06
07
19
19
08
Source: Anurio de Estadística Fiscal, 1912-13.
09
19
10
19
11
19
12
19
19
13
Bankers lent to themselves
Percent of non-government loans made to
banks’ own boards of directors:
Banamex 1886 to 1901
100%
Mercantil de Veracruz 1898-1906
86%
Coahuila, 1908
72%
Durango, 1908
51%
Mercantil de Monterrey, 1908
31%
Nuevo León, 1908
29%
Limits on bank entry created barriers
to entry in downstream industries
Four Firm Ratios in Textile Manufacturing
40%
35%
30%
25%
20%
15%
Mexico
10%
Mexico
Expected
Brazil
5%
USA
India
0%
1888
1891
1893
1895
1900
1904
1909
1912
1913
The experiment ended with a revolution—and
a bank expropriation using the supervisory
powers of the new government
Ratio of Bank Assets to GDP, Mexico, 1897-1929
35%
30%
25%
20%
15%
10%
5%
Haber, Razo, and Maurer 2003.
29
28
19
27
19
26
19
25
19
24
19
23
19
22
19
21
19
20
19
19
19
18
19
17
19
16
19
15
19
14
19
13
19
12
19
11
19
10
19
09
19
08
19
07
19
06
19
05
19
04
19
03
19
02
19
01
19
00
19
99
19
98
18
18
18
97
0%
Mexico’s second experiment:
Banking under the PRI
Getting bankers to deploy their capital meant:
A. Allowing them to write the banking laws
B. Limiting entry.
C. Giving them voice in the supervisory body—
the CNB.
D. Creating a government owned commercial
bank, Banxico, that made loans to the private
banks and to powerful politicians.
The lack of limits on government meant
that the rules were soon changed
In 1932 Banxico was made a central bank.
In 1936 the government required that private
banks hold deposit reserves in Banxico, and
bank supervision was transferred to
Banxico.
In the 1970s, the government used the supervisory
powers of Banxico to expropriate deposits, and then
the government expropriated the banks in 1982
Figure 2.8: Financial Repression in Mexico, 1948-82
60%
50%
40%
Reserve Ratio
Deposits/GDP
Private Credit/GDP
30%
20%
10%
0%
1948 1950 1952 1954 1956 1958 1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982
-10%
Source: Internatinoal Monetary Fund, International Finance Statistics.
Mexico’s Third Experiment: The
Privatization of 1991-96
1. Bankers faced expropriation risk.
2. Government wanted to maximize price on offer.
3. Government aligned the incentives of the bankers by
regulating entry
4. Bankers were also allowed to buy the banks with
borrowed funds—some of it from the same banks
they were buying.
5. Bankers therefore had weak incentives to be prudent,
which in turn required that depositors be protected
with unlimited deposit insurance, removing their
incentives to monitor.
So, Bankers Plunged into Credit Markets
Non-Government Bank Lending/GDP
0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
Made lots of bad loans, and got
bailed out
The Evolution of Non-Performing Loans, 1991-2003
60.0%
FOBAPROA Swaps
Rolled Over
Declared
As Percent of Total Loans
50.0%
40.0%
30.0%
20.0%
10.0%
0.0%
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Source: Haber 2004
Mexico’s Fourth Experiment: Bank Regulation in an
Emerging Democracy, 1997-2007
In 1997:
A. Accounting standards rewritten.
B. Related lending made more difficult.
C. Deposit insurance agency made independent.
D. Foreign banks permitted to buy Mexican banks.
In 2001-02:
A. Bankruptcy laws reformed;
B. Programs to improve property registers.
C. Expansion of non-bank banking sector.
In 2006:
A. Bank charters granted to large retailers.
Bank Lending to Households and
Business Enterprises, by type, as % GDP
16.0%
14.0%
12.0%
10.0%
SOFOL
Housing
8.0%
Consumer
Commercial
6.0%
4.0%
2.0%
0.0%
2000
2001
2002
2003
2004
2005
2006
2007
Conclusions
1. Broadly speaking, BCL’s main conclusions from cross
sectional evidence are consistent with case study.
2. When government’s authority and discretion were not
limited, supervision was used to either expropriate banks
(1916 & 1982), create government owned banks (19821991), behave opportunistically toward banks (1970s,
early 1990s).
3. When government’s authority and discretion were limited
(1997-2007), prudent supervision produced financial
development—but starting from a low base.
These results extend beyond Mexico
Time constraints prevent us from discussing other
cases.
Nevertheless, the analysis of other cases—Brazil,
Colombia, the United States—produces broadly
similar results
Supervision can be used opportunistically by
governments against private banks—and the more
unconstrained the government is, the more
opportunistic it can be.