Transcript Comments on - Inter-American Development Bank

Comments on:
Financial Development, Financial Fragility, and Growth
by Norman Loayza and Romain Ranciere
Graciela L. Kaminsky
George Washington University
Second Workshop of the Latin American Finance Network
Cartagena, Colombia
December 3, 2004
Motivation
•
Financial development triggers growth:
– The endogenous growth literature claims this link. For example, Ross Levine (2004)
reports that increasing financial deepening from the mean of the lowest quartile to
the mean of the upper quartile (of the distribution of domestic credit/GDP)
increases growth by 1 percentage point.
– Bekaert, Campbell and Lundblad ( 2001) results indicate that liberalization
promotes growth, with growth rates increasing 1 percentage point following
liberalization.
•
Financial liberalization harms growth
– The international finance literature suggests that financial liberalization triggers
“excesses” with booms and busts in financial markets and a collapse in economic
activity.
– Kaminsky and Reinhart (1999) results indicate that the probabilities of a banking
crisis increase by 40 percent following financial liberalization. See also, Detragiache
and Demirguc-Kunt (1999).
•
These opposite results have coexisted without any explanation.
Goals of the Paper
• This paper tries to explain the apparently contradictory results of the
literature.
• In particular, the authors try to untangle the short-run and long-run
effects of financial deepening on output growth.
• Short-run pain, long-run gain?
• Possible explanations.
The Estimations
p 1
q 1
j 1
j 0
 ( yi )t    ij  ( yi )t  j    ji  ( X i )t  j   '[( yi )t 1  {  oi   1i ( X i )t 1 }]   it
y = GDP per capita growth rate
X = control variables including:
financial depth
initial level of GDP
government consumption/GDP
volume of trade/GDP
inflation rate.
The Data
• Between 66 and 82 developing and developed countries,
depending on data availability.
• Sample: 1960-2000.
The Results
• The Long Run: A 1 percent increase in the ratio of private credit to
GDP leads to a rise of 0.7 percentage points in the growth rate of per
capital GDP.
• The Short Run: The average response of the growth rate to an increase
in the growth rate of private credit/GDP is -6.
• Interpretation: Boom-bust cycles in credit lead to a temporary output
collapse, still permanent financial deepening promotes growth.
The Short-Run
• Can the short-run adverse effect of credit on growth be explained by:
– Banking crises? (number of years of systemic banking crises in the
1960-2000 period)
– Financial volatility? (standard deviation of the growth rate of the
private credit to GDP ratio over the period 1960-2000).
• Yes
More Results
• Typical growth equations with a pooled (cross-country, time-series)
data set.
• Eighty two countries.
• For each country, 8 non-overlapping five-year periods over 1960-2000.
• Financial deepening has a positive effect on growth but …
• Banking crises and financial volatility harm growth.
Supporting Evidence on:
Short- Run Pain, Long-Run Gain
Average Amplitude of Booms and Crashes
(in percent)
130
120
110
100
90
80
70
60
50
40
30
20
10
0
All Markets
Booms
Repression
•
All Markets
Crashes
Emerging Markets
Booms
Short-Run Liberalization
Emerging Markets
Crashes
Mature Markets
Booms
Mature Markets
Crashes
Long-Run Liberalization
Source: “Short-Run Pain, Long-Run Gain: The Effects of Financial Liberalization” Graciela Kaminsky and Sergio Schmukler, NBER
Working Paper 9787, June 2003.
Figure 1
Index of Financial Liberalization
Less
3.5
Liberalization
Emerging Markets
3
2.5
2
1.5
Mature Markets
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
1979
1978
1977
1976
1975
1974
More
Liberalization 0.5
1973
1
Financial Liberalization and Institutional Reforms
Panel A
Sequencing
Mature Markets
Probabilities of Liberalization Conditional on
Insider Trading Insider Trading
Law and Order
Laws Existence Laws Enforcement
Type of Financial Liberalization
Partial Liberalization
Full Liberalization
36 **
64 ***
Hypothesis Test (P-Value)
Partial Liberalization = Full Liberalization
0.04
17
25 *
0.34
44 ***
50 ***
0.33
Emerging Markets
Type of Financial Liberalization
Partial Liberalization
Full Liberalization
Hypothesis Test (P-Value)
Partial Liberalization = Full Liberalization
Probabilities of Liberalization Conditional on
Insider Trading Insider Trading
Law and Order
Laws Existence Laws Enforcement
62 ***
77 ***
0.17
11
44 **
0.08
18
64 ***
0.02
Comments
•
Findings from the stock market responses to liberalization and the stories that
the authors tell in the theoretical section of the paper suggest that:
– Threshold levels may matter in the relationship between financial
development and growth. Booms in credit may not trigger collapses in
economic activity in mature financial markets.
– Do institutions matter when examining the short-run effect of financial
development? Perhaps interaction terms?
•
Also, is there a linear relationship between financial deepening and growth?
Isn’t  = 0.7 too big? Does financial depth affect the steady state growth
1
rate?
More Comments
•
In the long run, financial depth fuels growth but in the short run it harms
growth. What are the channels?
– Do short-run booms hurt capital accumulation? or productivity?
– Similarly, for the long-run relationship: Is it capital accumulation or
productivity?
•
Financial development goes hand in hand with financial liberalization (and
integration with world capital markets). But here Norman and Romain just
look at domestic financing not world financing. Physical location may not
matter as much with financial globalization. Bias???
•
Report in a graph the short-run responses by country (organized by size) with
the country names.
Last and Very Important Comment
• Kaminsky in the abstract of the paper should be written
with a final Y not i.
Conclusions
• Very interesting paper, a must-read.
• Very careful estimations.
• The first paper to look jointly at the short- and long-run effects
of financial deepening on growth. I like that section very much.
The authors should expand this part of the paper.