Transcript Trade integration and risk sharing in Sub

Trade integration and risk sharing
in Sub-Saharan Africa
Do Institutions and Financial
Depth Matter?
By
John Bosco Nnyanzi
Supervisors:
Prof. Dr. Michael Landesmann
Prof. Dr. Joseph Francois
20/07/2015
1
Introduction – Research Question





Investments in markets of other countries
help reduce exposure to idiosyncratic risks
arising from country-specific shocks.
Via diversification of portfolios
This is called risk sharing (RS)
Studies show low RS in developing countries
(Kose, et al., 2009)
But precise avenues via which risk is shared
are not unambiguous
20/07/2015
2
Four Issues
1.
2.
3.
4.
Extent to which trade integration contributes
to consumption risk sharing;
The role of institutional quality to
consumption Risk Sharing;
The extent to which financial depth affects
consumption risk sharing ; and,
The extent to which the role of trade
integration to risk sharing depends on
institutional quality and/or financial depth.
20/07/2015
3
Role of Integration?













Focus: EAC, ECOWAS, SADC, and AU, - 1986-2007.
Why these groups?
Established in Africa during the 1990s with an aim
to expand trade via trade integration,
increase capital and financial flows,
impact on risk sharing is still an empirical question.
By trade integration we mean trade openness (TOP)
Why?
economic agents forge economic ties,
flow of capital,
joint production and development,
transaction costs
Foreign bank participation
20/07/2015
4
Trade cont’d

Monetary union: removal of trade restrictions as pre-condition. Why?

Trade integration helps in greater synchronization of economic cycles
(United Nations Economic Commission for Africa, 2008).

In absence of access to international capital markets,
countries, via trade openness, could enjoy the risk sharing benefits from
international integration.


The expected benefits of risk sharing accruing from TOP have however
proven hard to substantiate, and its impact on risk sharing is still an
empirical question.

TOP contribution could depend on domestic financial markets and
institutional environment (Broner and Ventura, 2006).

Does extent to which TOP contributes to risk sharing depend on institutional
quality and financial depth?
20/07/2015
5
Why Financial depth?





Financial depth (proxied by credit to GDP)
Access to capital
for investment projects that investors might forego;
facilitate trade.
Determine volume of trade across borders.
NOTE:
 Financial deepening is much less developed in SSA than in most of
the emerging market countries.
 Domestic credit to private sector as percentage of GDP in SSA
decreased from 69.3% to 58.5% in the year 2007 and 2008
respectively.
 Implication of trend to risk sharing still an empirical question.

Does the potentiality of TOP to influence CRS depend on how it acts
as proxy for a country’s creditworthiness?
20/07/2015
6
Why institutions?

Matter for investment decisions

Incentive to trade e.g. the protection of property rights (Olson, 1996)
Encourage contract enforcement and trust (Dixit, 2009; Volosovych,
2005)







higher rates of return via lowering transaction costs.
Corruption and bureaucracy increase transaction costs and
discourage investment (Papaioannou, 2009; Wei and Wu, 2002)
Outward flow impact:
E.g, capital flight. How??
increasing the riskiness of the economy or
depressing the domestic investment climate (Cerra, et al., 2008;
Ali, 2010).
20/07/2015
7
Literature









Broner and Ventura (2006): Effects of TOP depend on domestic
financial markets.
If deep, trade TOP allows for better risk-sharing;
if thin, trade integration destroys risk sharing.
Giovanni et al., (2006): more open to trade countries tend to be
more volatile,
contradicts TOP as shock absorber (Martin et al., 2006).
Kose et al., (2007): No substantial evidence in in emerging
economies.
role of financial integration in risk sharing depends on TOP.
Ventura (2008) and Corcoran (2007): role of trade openness
appears possible.
So, role of trade integration for risk sharing is imprecise.
20/07/2015
8
Literature Cont’d






Kose et al.,(2007), do not find institutions and credit
depth helpful in risk sharing
but call for more careful investigation.
Fratzscher et al., (2009): low institutional quality but
high degree of risk sharing if financially open.
closed economies experience virtually no risk sharing.
high levels of investor protection leads to less share
of consumption risk (Scharler, 2004)
potential risk sharing gains are smaller in countries
with better perceived institutional quality (Imbs and
Mauro, 2007)
20/07/2015
9
Model

Practically we adopt a model similar to that of ASY (1996)
that takes the form:
 log Cit   log Ct  i   ( log GDPit   log GDPt )  it




is the year-on-year growth rate of real consumption
per capita for country i in year t;
 log C
is the growth rate for “world” real consumption per
capita;
 log GDP is the year-on-year growth rate of real GDP per
country i in year t;
 logGDP is the growth rate of the “world” real GDP;
 log C
it
t
20/07/2015
it
10
Model cont’d




measures the average co-movement of the
countries’ idiosyncratic consumption growth with
their idiosyncratic GDP growth during the entire
time period.
Its slope measures the average deviation from
perfect risk sharing in consumption.
100(1   )%
A
measures the degree of
international risk sharing in percentage terms.
20/07/2015
11
Interaction-based Model

We follow Sørensen, Wu, Yosha, and Zhu (2007)
by in addition allowing  to change over time so
that
  0  1 (t  t )  2 ( X it  X t )



Xit is proxy for trade openness and/or, domestic
institutional factors or financial depth.
X t is the average of Xit across countries.
t t
is a time trend that captures the trend
decline/increase in risk sharing not directly caused
by the potential variables.
20/07/2015
12
Model cont’d

Therefore we estimate the following
parametized regression model
 log Cit   log Ct  i  0 (t  t )  1 ( X it  X t )


  2  3[t  t ]   4 [ X it  X t ]   log GDPit   log GDPt 
 it
20/07/2015
13
Data










Study covers Africa with data for period 1986-2007
Data on real per capita consumption and real per capita GDP
measured in real constant terms are from PWT 7.0 (2011).
Data on TOP (proxied by the sum of exports and imports over GDP)
is from PWT 7.0.
Data on financial depth (proxied by the ratio of total credits to GDP)
is from World Development Indicators (WDI).
Data on institutional quality is from PRS
Control for financial integration (measured as sum of a country’s
foreign assets and liabilities to GDP)
Data is from External Wealth of Nations Mark II by Milesi-Ferreti
(2008) and WDI.
Control for changes in real exchange rate.
Data on this variable is from PWT version 7.0.
Calculate the world aggregates as GDP weighted averages.
20/07/2015
14
Table 1 Descriptive Statistics
20/07/2015
15
Table 1 Cont’d
20/07/2015
16
Table 2 Correlation Matrix
20/07/2015
17
20/07/2015
18
20/07/2015
19
20/07/2015
20
20/07/2015
21
20/07/2015
22
20/07/2015
23
Summary of Results











SSA experiences about 34% CRS – Table 3a
TOP has a positive impact on CRS – Table 3a
Indications that Financial depth helps in risk sharing but weak
evidence – Table 4a
Weak evidence that aggregate institutions matter for CRS –
Table 5a
Individual Institutions matter: Tables 5a & 5b
Corruption in SSA & ECOWAS & AU
Government Stability in EAC & SADC
Bureaucracy, Rule of law, in SADC & AU
Institutions could be deleterious e.g. Bureaucracy in SADC & AU
TOP contribution could depend on Institutional quality and
financial depth – Tables 6 &7
FOP and CRE in interaction facilitate TOP role in CRS – Table 8
20/07/2015
24
End

Thanks for listening
20/07/2015
25