Transcript The Remittances impact on Financial Development

Academy of Economic Studies, Bucharest
Doctoral School of Finance and Banking
Impact of Remittances on
Financial Development and
Economic Growth
Supervisor:
Professor Moisă Altăr
Student:
Nita Olivia Georgiana
Structure:

Introduction

Literature review

Methodology and data

Empirical results

Conclusions

Bibliography
Introduction

Remittances - transfers of resources from individuals in one country to
individuals in another - are an important source of private funds in
developing countries.

Remittances have grown from U.S. $31.2 billion in 1990 to U.S. $338 billion
in 2008.1

Migrants from developing countries sent home $316 billion during 2009 or
6% less than they did during the previous year. The World Bank forecasts a
6.2% increase in remittances this year.

The research subject: remittances and their impact on growth and financial
development.

The main goal of the research: the theoretical and methodological study of
remittances, the economic analysis and the use of remittances, the
importance of remittances in promoting growth, by looking at the interaction
between remittances and financial sector and also the impact of remittances
on financial development.
1
The Economist, Apr 29th 2010, “Remittances”
Literature review

More than 60% of remittances are used to purchase daily necessities such
as food, clothes and shelter, remittances are a key poverty reduction tool
(Adams and Page 2005; Acosta et al. 2008).

The 20-40% of remittances that is used to save or invest is the key to
achieving a family’s longer-term financial independence.

IMF surveys have shown that 30-50% of remittances recipients have
access to a bank account. To expand this area, banks can offer a greater
range of financial products such as: microcredit, insurance and remittancebacked mortgages.

The fact that remittances might affect financial development in developing
countries is based on the concept that money transferred through financial
institutions give access for recipients to other financial products and
services, which they might not have otherwise (Orozco and Fedewa, 2005).
Economic Models using Remittances


The Keynesian Model
The equilibrium condition is:
Y  D
DCI
Y C S
S  I 
For an open economy equilibrium
condition must be supplemented with the
influence of external transactions:
YM DX
S M  I  X

The Keynesian Model
Variable used: the effective demand
for goods (D), total final
consumption (C), savings (S), global
investment (I), income or production
(Y), imports (M) and exports (X) .
An increase in Y due to the increase
in the remittances R flows can be
represented either by an independent
increase of exports receipts, either
through additional investment.
An additional inflow in R will
increase the income balance from
point A to point B.
The final income balance will
depend essentially on the R
influence on the propensity to
saving and consumption of imports.
The IS-LM Model



The IS-LM Model
Variable used: the real sector (IS),
the monetary sector (LM) and the
external one (BP) are in
equilibrium (point E), at a given
level of income Y1 and a certain
interest rate i1.
BP curve is perfectly inelastic.
Forced by remittances flow, the
monetary expansion will lead to
increased revenue ( Y 2), which, in
turn, will condition the cheaper
domestic credit as the real sector
will be growing.
The National Accounts System


The additional remittance flows increases the aggregate demand and is
integrated in the gross national income available.
This can be expressed as follows:
Y  ( C  I ) p  ( C  I ) g  Y f  Tr f  ( X  M )
- gross national income;
( C  I ) p - consumption and private investment;
( C  I ) g - consumption and government investment;
- foreign income;
Yf
- net current transfers from abroad (remittances);
Tr f
X
- exports of goods and services;
M
- import of goods and services.
The available gross national income contains the current account of the
balance of payments CAB  ( X  M )  Y  Tr
Transmitted in the country, these resources can be saved, consumed or
invested.
 Y







f

f
Methodology and data

This paper uses balance of payments data on remittance flows received by
10 European developing countries. I worked with a panel data using
Eviews 7 econometric tools.

I start by estimating the impact of remittances on economic growth by
ordinary last squares (OLS), without using any financial development
variables:
L _ GDP i ,t   1 Re m i ,t   2 X i ,t   i  u i ,t
where i refers to the country and t to the time period from 1994 to 2009

L_GDP, denotes the logarithm of initial level of GDP per capita;

Rem refers to the ratio of remittances to GDP. The data on remittances are
constructed as the sum of three items in the Balance of Payments Statistics
Yearbook (IMF): workers’ remittances, compensation of employees and
migrant transfers.
Figure1
 Figure1 shows the European remittance recipient countries used
in this sample for the year 2009, measured as a percentage of GDP.
 Moldova (18.16%), Albania (15.22%), Serbia (9.81%), Bulgaria
(3.66%) and Romania (1.64%) are among the largest recipients of
remittances as percentage of GDP .

Matrix X refers to a set of control variables that the literature has found to
affect economic growth and financial development:

Inflation, measured as the annual percentage change in the consumption
price index;

Openness to international trade, defined as the ratio of the sum of exports
plus imports of goods to total output;

Other flows to GDP measured as the ratio of capital inflows to GDP
(including aid, FDI, and portfolio flows), and

Population growth.

First I analyzed the relationship between remittances and economic growth
by running fixed effects and random effects regressions, ignoring the
potential for biases due to reverse causation or measurement error.

Hausman test is reported for comparing the efficiency of random and fixed
effect estimates.

To address the endogeneity problem the Generalized Method of Moments
(GMM) panel data estimator, developed by Arellano, M. and Bover, O.
(1995) is used for estimations with lagged regressors as instruments.

Regression estimated with the GMM method:
L _ GDP i , t   0   1 L _ GDP i ,t 1   2 Re m i ,t   3 FD i ,t   4 Re m i ,t * FD it    5 X it   i  u i ,t

To explore the relationship between financial development and remittances
the following equation is estimated:
FD i , t   1 Re m i , t   2 X

i ,t
  i  u i ,t
FD, financial development, refers either to the ratio of bank credit to the
private sector or the share of bank deposits expressed as a percentage of
GDP.
Empirical results
Table 3
Panel estimates of the Impact of Remittances on Economic
Growth
Fixed Effects Results (OLS)
Log of GDP
Remittances to GDP
Openness
Other flows
Population growth
Inflation
Constant
Observations
Number of countries
0.011
[8.55]***
0.007
[15.24***
0.001
[2.08]***
0.169
[12.27]***
-0.003
[-6.73]***
7.031
[153.16]***
150
10
Absolute values of t statistics are in brackets. The symbols *, ** and ***
denote significance at the 10, 5 and 1 percent level, respectively.




In Table 3 we can find the
fixed effects estimates.
The
relationship
between
remittances and growth is a
positive one.
A one percentage point
increase in the share of
remittances to GDP suggests a
0.011
percentage
point
increase in the economic
growth.
The economic growth is
positively influenced by all
variables,
but
negatively
influenced by inflation.
Table 4
Panel estimates of the Impact of Remittances on Economic Growth
Random Effects Results (OLS)
Log of GDP
Remittances to GDP
Observations
0.010
[3.82]***
0.006
[7.82]***
0.003
[2.10]**
0.189
[8.64]***
-0.002
[-3.81]***
7.016
[57.35]***
142
Number of countries
10
Hausman test
P-value for Hausman test
52.65
0.00
Openness
Other flows
Population growth
Inflation
Constant




Absolute values of t statistics are in brackets. The symbols *, ** and
*** denote significance at the 10, 5 and 1 percent level, respectively.
Using the random effects
estimates, we also find that the
relationship between remittances
and growth is a positive one.
A one percentage point increase
in the share of remittances to GDP
suggests a 0.010 percentage point
increase in the economic growth.
The
economic
growth
is
positively influenced by all
variables,
but
negatively
influenced by inflation.
The Hausman test shows that the
fixed effects model is preferable.
Table 5
GMM Panel Data estimates of Remittances, Financial
Development and Economic Growth
Remittances to GDP
Bank Deposits to GDP
RemGDP*BankDepGDP
Log of GDP
Log of GDP
0.010
[3.29]***
0.002
[3.66]***
-0.0001
[-1.92]***
0.005
[2.31]**
0.002
[4.05]***
-0.001
[-3.25]***
0.012
[0.46]
-0.001
[-1.86]*
0.760
[12.23]***
0.001
[2.19]**
-2.08
[-0.37]
0.002
[3.46]***
-0.002
[-2.33]**
0.046
[2.14]**
-0.001
[-2.65]**
0.719
[8.29]***
102
1.50
116
0.90
0.47
0.82
Bank Credit to GDP
RemGDP*BankCreditGDP
Openness
Other flows
Population growth
Inflation
Lag 1 of Log of GDP
Observations
Sargan test for
overidentifying restrictions
P-value Sargan test
Absolute values of t statistics are in brackets. The symbols *, ** and ***
denote significance at the 10, 5 and 1 percent level, respectively.
 To address the endogeneity
problem I used the Generalized
Method of Moments (GMM),
following Arellano and Bover (1995).
 In order to ensure that the
interaction term does not proxy for
remittances or the level of
development of financial markets,
these variables are also included in
the regression separately.
Growth is positively influenced by
both remittances and financial
development.
 The interaction sign is negative,
which indicates that remittances and
financial development are used as
substitutes to promote growth.
The Sargan p-value shows the
validity of the instruments, the null
hypothesis cannot be rejected .
Table 6
Panel estimates of the Impact of Remittances on Financial
Development
Fixed Effects Estimates (OLS)
Remittances to GDP
GDP per capita
Other flows
Population growth
Inflation
Openness
Constant
Bank Deposits to
GDP
Bank Credit to GDP
0.770
[13.28]***
2.935
[3.29]***
0.163
[3.88]***
9.126
[4.45]***
-0.039
[-1.86]*
-0.023
[-0.78]
0.718
[10.48]***
7.529
[7.28]***
0.664
[7.20]***
4.973
[3.805]***
-0.062
[-2.25]**
-0.114
[-2.29]**
29.37
[15.93]***
13.93
[4.27]***
Absolute values of t statistics are in brackets. The symbols *, ** and ***
denote significance at the 10, 5 and 1 percent level, respectively.



The next step now is to analyze the
remittances direct impact on
financial development.
As we can see the remittances have
a positive coefficient for both
measures of financial development.
Assuming a causal relationship, a
one percentage point increase in the
share of remittances suggests a 0.77
percentage point increase in the
ratio of deposits to GDP, while it
leads to a 0.71 percentage point rise
in the share of credit to GDP.
Forecast 2010-2011
20
Forecast: D_BCF
Actual: D_BC
Forecast sample: 1994 2011
Adjusted sample: 1995 2011
Included observations: 139
Root Mean Squared Error 6.956537
Mean Absolute Error
3.938316
Mean Abs. Percent Error
177.0869
Theil Inequality Coefficient 0.778159
Bias Proportion
0.000000
Variance Proportion
0.896889
Covariance Proportion
0.103111
16
12
8
4
0
-4
-8
-12

Albania Albania Albania Bulgaria Bulgaria Bulgaria Czech Republic Czech Republic Czech Republic Hungary Hungary Moldova Moldova Moldova Poland Poland Poland Romania Romania Romania Serbia Serbia Slovak Republic Slovak Republic Slovak Republic Ukraine Ukraine -
95
01
07
98
04
10
99
05
11
01
07
97
03
09
98
04
10
99
05
11
03
09
98
04
10
01
07
-16
D_BCF

± 2 S.E.
Figure 2
30
20
10
0
-10
-20
-30
-40
-50
Albania - 94
Albania - 00
Albania - 06
Albania - 12
Bulgaria - 99
Bulgaria - 05
Bulgaria - 11
Czech Republic - 98
Czech Republic - 04
Czech Republic - 10
Hungary - 97
Hungary - 03
Hungary - 09
Moldova - 96
Moldova - 02
Moldova - 08
Poland - 95
Poland - 01
Poland - 07
Romania - 94
Romania - 00
Romania - 06
Romania - 12
Serbia - 99
Serbia - 05
Serbia - 11
Slovak Republic - 98
Slovak Republic - 04
Slovak Republic - 10
Ukraine - 97
Ukraine - 03
Ukraine - 09
-60
D_BC
Figure 3
D_BCF

Remittance flows are broadly
affected by three factors: the
migrant stocks, incomes of
migrants in the destination country
and incomes in the source country.
Figure 2 reports the results of the
forecast for the ratio of credit to
GDP and Figure 3 presents the
differences between the two series
actual and fitted for the ratio of
credit to GDP by analyzing the line
graph.
A 1% fall in remittances suggests a
fall of about 0.23% in the ratio of
the credit to GDP.

12
Forecast: D_BDF1
Actual: D_BD
Forecast sample: 1994 2011
Adjusted sample: 1994 2011
Included observations: 135
Root Mean Squared Error 4.002412
Mean Absolute Error
2.438890
Mean Abs. Percent Error
169.8499
Theil Inequality Coefficient 0.730591
Bias Proportion
0.000000
Variance Proportion
0.875672
Covariance Proportion 0.124328
8
4
0
-4
Albania Albania Albania Bulgaria Bulgaria Bulgaria Czech Republic Czech Republic Czech Republic Hungary Hungary Hungary Moldova Moldova Moldova Poland Poland Poland Romania Romania Romania Serbia Serbia Slovak Republic Slovak Republic Slovak Republic Ukraine Ukraine Ukraine -
94
00
06
96
02
08
96
02
08
97
03
09
98
04
10
98
04
10
98
04
10
01
07
95
01
07
97
03
09
-8
D_BDF1

± 2 S.E.
Figure 4 reports the results of the
forecast for the ratio of deposits to
GDP and Figure 5 presents the
differences between the two series
actual and fitted for the ratio of
credit to GDP by analyzing the line
graph.
A 1% fall in remittances suggests a
fall of about 0.65% in the ratio of
the deposits to GDP.
Figure 4

30
20
10
0

-10
Albania - 94
Albania - 00
Albania - 06
Albania - 12
Bulgaria - 99
Bulgaria - 05
Bulgaria - 11
Czech Republic - 98
Czech Republic - 04
Czech Republic - 10
Hungary - 97
Hungary - 03
Hungary - 09
Moldova - 96
Moldova - 02
Moldova - 08
Poland - 95
Poland - 01
Poland - 07
Romania - 94
Romania - 00
Romania - 06
Romania - 12
Serbia - 99
Serbia - 05
Serbia - 11
Slovak Republic - 98
Slovak Republic - 04
Slovak Republic - 10
Ukraine - 97
Ukraine - 03
Ukraine - 09
-20
D_BD
Figure 5
D_BDF
From these forecasts we continue to
find that remittances are an
important source of external funds.
That’s why migrants should be
stimulated to continue to send
remittances home which will bring
economic growth and will also have
a positive influence on financial
development.
Factors affecting remittances flows in 2009




Effect of current crisis on migration stocks and flow
Remittance flows in a given year are not directly related to migration flows
during the same year; remittances are sent by almost the entire existing stock
of migrants.
The return migration is as a result of the financial crisis in the US and
Europe and the new migration flows, which have been impacted by the
financial crisis and weak job markets in the destination countries.
Efforts by migrants to cut consumption
Remittances are a small share of migrants’ incomes, and they typically
continue to send remittances even when hit by income shocks.
Currency effects
An important factor affecting the currency valuation of remittances is the
change in the exchange rates between the relevant local currency and the
remittance’ s currency. Exchange rate changes also appear to affect the
consumption/investment motivation for remittances.
Conclusions

Remittances promote growth in less financially developing countries by
providing an alternative way to finance investment.

Remittances acted as substitutes for financial services in promoting growth,
by offering the response to the credit needs and insurance that the market
has failed to provide.

The fact that remittances contribute to overcome liquidity constraints and
help undertake profitable investment in developing countries is important
for future research.

Increasing the official inflow of remittances

Migrant transfers should be stimulated; the role played by migrants should
be recognized and reinforced.

A low–cost and secure remittances transfer service should be provided.

Strategic policies should be combined with measures to encourage the
transfer and investment of remittances to promote economic growth.



One way of bringing more remittances and increasing financial
development implies that banks reconsider the conditions for bank credits
and deposits.
Banks can make themselves and their services appealing to remittances
senders and receivers.
Some of the policies that banks can implement are:




include the remittances received when calculating the income in order
to determine creditworthiness;
implement mechanisms to channel remittances directly and
conveniently to financial products or regular expenses;
market financial products to remittance senders and receivers;
facilitates remittance withdrawals and deposits and develop financial
products targeted specifically at the remittance market.
Selected Bibliography










Paola Giuliano, Marta Ruiz- Arranz (2009), “Remittances, financial development, and
growth”, Journal of Development Economics 90, 144-152.
Reena Aggarwal, Asli Demigruc-Kunt, Maria Soledad Martinez Peria (June 2006), “Do
Workers’ Remittances Promote Financial Development”, The World Bank.
Thomas H.W. Ziesemer (2010), “The impact of the credit crisis on poor developing countries:
Growth, worker remittances, accumulation and migration”, Economic Modelling.
Richard H. Adams JR (2007), “The Determinants of International Remittances in Developing
Countries”, World Bank, Washington, DC, USA.
Dilip Ratha, Sanket Mohapatra (2007), “Increasing the Macroeconomic Impact of
Remittances on Development”, The World Bank, Development Prospects Group.
Conrad Heilmann (2006), “Remittances and the migration–development nexus—Challenges
for the sustainable governance of migration”, Ecological Economics, 231-236.
Caroline Freund, Nikola Spatafora (2008), “Remittances, transaction costs, and informality”,
Journal of Development Economics 86,356–366.
Bradford Barham, Stephan Boucher, (1998), “Migration, remittances, and inequality:
estimating the net effects of migration on income distribution”, Journal of Development
Economics, vol. 55, 307-331.
Denise Stanley, Radha Bhattacharya (2008), “The informal financial sector in the U.S.: The
role of remittances”, The Quarterly Review of Economics and Finance, 48,1–21.
Arellano, Manuel and Olympia Bover, “Another Look at the Instrumental Variable Estimation
of Error-components Models”, Journal of Econometrics 68, 29-51, 1995
Thank you for your time!
Back up slides

The Hausman Test

Hypothesis:

If H 0 is true, both fixed effects
estimator and random effects
estimator are consistent, but only
the random one is efficient.
If H a is true, the fixed effects
estimator is consistent and the
random one is not.
In this case the null hypothesis is
rejected, the fixed effect estimator
is preferable.
Correlated Random Effects - Hausman Test
Equation: GDP_RE
Test cross-section random effects
Test Summary
Cross-section random
Chi-Sq. Chi-Sq.
Statistic
d.f.
52.654297
5
Prob.
0.0000



The Sargan Test

Sargan test is used for over identifying restrictions.

Under the null hypothesis that the over-identifying restrictions are valid,
the Sargan statistic (J-statistic) is distributed  ( p  k ) ,where k is the
number of estimated coefficients and p is the instrument rank.

The p-values show that the validity of the instruments ,the null hypothesis
can not be rejected .

The high p-values 0.47, 0.82 respectively suggest almost certain acceptance
of the null hypothesis , that the variables are valid.