Transcript Presentation_Christodoulakis_Sarantides

External asymmetries in the Euro-Area
and the role of Foreign Direct Investment
Nicos Christodoulakis
Vassilis Sarantides
Athens University of Economics and Business
January 2011
1
It is by now well recognized that large
CA asymmetries between East-West were responsible
for the 2008 credit crunch
Current Account, %GDP
China
US
8
3
-2
2006
2007
2008
2005
2003
2004
2001
2002
1999
2000
1997
1998
1996
1993
1994
1995
1992
1990
1991
1988
1989
1986
1987
1984
1985
1983
1980
1981
1982
-7
2
After the credit crunch,
alarming spreads in the Eurozone!
SPR EA D O F 10Y B O ND OVER THE B U ND
( m o n t h l y d a ta i n B P S )
300
250
E L
IE
IT
200
150
P T
S P
B E
100
50
0
G lo b a l
c ri s is
-5 0
01
02
03
04
05
06
07
08
09
But not just for the highly indebted countries…
3
Conventional wisdom:
The cost of borrowing increases with indebtedness
120
IT
110
GR
BEL
Public Debt (%of GDP)
100
90
80
FR
70
AT
PT
NL
60
no GR
SP
FL
50
IE
(corr = +0.53)
40
30
20
5
15
25
35
45
Sovereign Spreads
55
Correlation between Public Debt and Spreads
(10y bond yield over the Bund, averages 1999-2009)
65
75
4
Un-conventional wisdom:
The cost of borrowing increases also with CA Deficits
)
8
6
current account (% of GDP
4
2
FL
NL
(corr = - 0.68)
BEL
AT
FR
0
-2 5
-4
45
35
25
15
55
75
65
SP
no GR
-6
-8
IT
IE
PT
GR
-10
-12
Sove re ign Spre ads
Correlation between Current Account and Spreads
(10y bond yield over the Bund, averages 1999-2009)
5
Voices of concern too late:
When the global crisis hit mostly
countries with Public and/or CA Deficits
• R. Zoellick (World Bank, 10 Oct. 2008)
…Emerging economies with large CA Deficits will be
hit by scarcity of international capital …
• UN (R. Shelbourne, June 2008)
… the tightening of credit … can turn the problem of
illiquidity into a problem of insolvency ..
6
TRADE BALANCE (% GDP), Source: IMF
COUNTRY
1980-1998
1999-2009
CHANGE
Austria
-0.114
3.061
2. 947
Finland
2.373
7.352
4. 979
Germany
1.763
3.986
2. 223
Ireland
3.830
12.878
9. 048
Netherlands
4.052
6.471
2. 419
North (average)
2.380
6.496
4.368
France
0.920
0.354
- 0. 566
Greece
-6.085
-8.330
- 2. 245
Italy
0.934
0.505
- 0. 429
Portugal
-8.492
-8.495
- 0. 003
Spain
-0.850
-3.856
- 3. 006
South (average)
-2.715
-3.964
- 1.250
7
TWO GROUPS IN EUROZONE
External Surplus, more FDI :
AT, BE, LX, FI, IE, NE, GE
“NORTH”
External Deficit, less FDI:
SP, PT, EL, FR, IT
“SOUTH”
8
Why External balances are widening after EMU?
Early theory: Demand-induced
Fall in real interest rates
 Higher consumption
 External imbalances, only temporary
… but SGP would reduce fiscal deficits
Investment would rise in the Eurozone,
and external balances would be corrected.
9
Investment has risen, but …intensity differed: Inward FDI much
more in the “NORTH” (exc. Belgium)
FDI (% GDP), Source: IMF
COUNTRY
1980-1998
1999-2009
CHANGE
Austria
0.644
5.644
5.000
Finland
0.976
3.125
2.149
Germany
0.255
2.459
2.204
Ireland
2.062
7.82
5.758
Netherlands
2.407
7.176
4.769
North (average)
1.268
5.244
3.976
France
0.969
2.994
2.025
Greece
1.074
0.844
-0.230
Italy
0.307
1.228
0.921
Portugal
1.609
2.766
1.157
Spain
1.615
3.634
2.019
South (average)
1.115
2.293
1.178
10
Three questions are looming:
1. Why the ‘North’ has attracted more FDI,
contrary to conventional predictions about
capital migration from rich to poor?
2. Why the composition Manufacturing/R-Estate
differed between North and South ?
3. Has FDI affected External Deficits and how ?
11
What can we learn from economic theory?
Investment in an open economy with traded and non-traded goods
e.g. Brock (1989), Turnovsky(1996)
Three factors of production
K : Internationally mobile capital: K   K  I
H : Immobile stock (Land, SMEs, etc) :
H Y
N
C
N
 cost( I )
Labour : fixed
Wages equalised across sectors
12
Internationally mobile capital
K
Trade Balance:
TB  YT  CT  I
T
K
FDI
N
T raded G oods
Non-traded Goods
YT  A  K T H T1
Y N  B  K N H 1N 
H
T
H
N
Immobile stock (Land, SMEs, etc)
13
Assumptions in the two-sector model
1. Perfect factor mobility between traded and non-traded production
Rental prices of each factor equalized across sectors.
RK 
YT
Y
p N
KT
K N
RH 
YT
Y
p N
H T
H N
2. Installing FDI entails non-traded costs (Hayasi)
cost( I )    ( I / K ) 2

I2
H  Y N  C N  c ost( I )  K  Y N  C N  
K
14
3. In EZ-South, quality of the Regulatory Environment
substantially inferior to that in EZ-North countries
Score in Regulatory Quality (max 2)
Average 1996-2008, Source: World Bank
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
AT
FI
GE
IE
NL
FR
EL
IT
PT
15
SP
To account for Regulatory impediments in the South,
modify production function in the Traded sector
3. There are obstructions in productive expansion due to
regulatory impediments, scarcity of industrial land, et c
x
 KT 

1
YT  

A

K
H


T
T
 HT 
A  KT  x H T1    x
For typical values α = 0.70, β = 0.30
North: x = (2 - 1.56)/2 = 0.21  α-x > β
South: x = (2 - 1.02)/2 = 0.49  α-x < β
16
The intertemporal problem
Choose :
CT , CN , I 
 max Ut 

1

1
1
s0 1   [CT (t  s)  CN (t  s)]  exp[ s]ds
Wealth abroad (B) with an exogenous world interest rate (r)
T r a d e b a la n c e :
T B  YT  C T  I

B  rB  T B  rB  YT  C T  I

C u r r e n t A c c o u n t:
CA  B
17
Hamiltonian:
 

1
[C
1  
 B

T
C
1
N
]1  
 (Q  )  K  ( p  )  H
1
 
[ C T  C N1   ]1     [ rB  Y T  C T  I ]
1
 ( Q  )  [   K  I ]  ( p  )  [Y N  C N
I2

]
K
18
First-order conditions:

0
CT

    

0
C N

B

(Q) (Q)  
K



  r




( p ) ( p )   
H


0
I

I 2
Q  (r   )Q  R K  p ( )
K


p  rp  R H
19
Solving the two-sector model
Consumption rule
Investment rule
C   r 1 



C


I
K
2 p
where   inflation=(1-  )
p
p
(Q  1)
Asset price, Q
1  x
2

I2

I
Q  (r   )Q  R K  p     (r   )Q  Z K  p    x  (
) p
K
K

Relative price, p

x
 x
p  r  p  RH  r  p  ZH  p
20
Returns determined from model parameters
(complicated expressions)
R K  ZK  p

1  x
   x
R H  ZH  p

 1    ( ) 
  
ZK  
1   
 
ZH   
  
 
  

(1  )(1  )
  
(1   )
  (1  )
  
(1   )
 (1  )
  


 x
   x
(1  )
  




  

(1  )
  
    x

  
21
Represent EMU as a permanent fall
in real interest rates
REAL LONGTERM INTEREST RATES
Source : AM ECO, 2010
1992-98 avg=4.84%
7
1999-07 avg=2.07%
6
5
4
3
2
1
0
BEL
GER
IRL
G RE
ESP
FRA
ITA
NDL
AUS
PRT
FIN
Fact: After EMU, real interest rates fell about -2.7 p.u. 22
Price equilibrium and Response to Shocks (I)
Traded sector intensive in FDI, β < α- x
ratio Q
E1
Q1
saddle path

new Q  0
Q0
E0


new p  0

p0
0
p1
Q0
p0
for small δ
relative prices
A permanent fall in interest rates
 asset price Q rises  Investment rises
Prices of non-traded goods fall relative to traded
23
Price equilibrium and Response to Shocks (II)
β>α-x
Non-traded sector Intensive in FDI,

new p  0
Q

newQ  0
Q2

p0
E2

Q0
Q0
0
E0
p0
p2
p
A permanent fall in interest rates
Asset price Q jumps , but only temporarily,
 Investment rises, but only for a while
Prices of non-traded goods rise relative to traded
24
Stock equilibrium and Response to Shocks
Stock-adjustment processes for stocks B and H
(1   )  RK
  RH
B  rB 
K 
H  C  I 
  x
  x
  RH
[(1   )  RK   ]
 C *
equilibrium: B* 
H *
K
r (    x )
r (    x )
r

I2
(1   x)  RK
(  x)  RH
I2
H  YN  CN  

K
H  CN  
K
(    x) p
(    x) p
K
(1   x)  RK
equilibrium: H* 
K  (  x   )   (1 )  C *  2K]
25
(  x)  RH
Equilibrium and Response to Shocks (III)

B
H 0

B0
B1
A permanent fall
in interest rates
B0
0
H0
H1
α-x>β, Productive sector intensive
H
in FDI,
26

B
A permanent fall
in interest rates
B0

H 0
B0
B2
0
H0
H2
H
α-x<β, Non-traded sector intensive in FDI
27
Stylised Facts of the Model
Proposition 1: Asset prices rise more in the North,
Q1
North
Q2
South
Q0
0
to
time
28
Fact 1: Asset prices higher in the North
After EMU, Ratio average Qnorth/Qsouth > 1
1.30
1.25
1.20
1.15
1.10
1.05
Q_NORTH/Q_S OUTH
1.00
0.95
2 00 0
2001
2002
2003
2004
2 00 5
2006
2007
2008
29
Proposition 2: Prices of Non-traded Goods rise more in the South:
If Traded sector FDI intensive: Prices of Non-traded DOWN
If Non-traded sector FDI intensive:
Prices of Non-traded UP
p
South
p2
productivity
p0
gap=p2-p1
North
p1
0
to
time
30
Fact 2: Post-EMU Inflation higher in the SOUTH
IN F L A T I O N R A T E S A F T E R E M U
(H IC P , E u r o s t a t )
3.2
3.0
2.8
2.6
2.4
2.2
2.0
1.8
1.6
2000
2001
2002
2003
A V G N O RTH
2004
2005
2006
2007
AV G S O UT H
Productivity Gap is accumulated
31
Proposition 3: As a result, FDI inflows are
expected to be higher in the North
Investment rule
I
K
2 p
(Q  1)
Compare FDI flows
N
FDI north
K
 S
FDI south
K
>1

 N

S
>1

S
p
N
p
>1
Q 1
 S
Q 1
N
>1
32

Fact 3: FDI inflows as % of GDP
Substantially higher in the North
EMU
14.00
12.00
Source: IMF
EXCLUDING BELGIUM
10.00
8.00
6.00
4.00
2.00
0.00
-2.001980
82
19
84
19
86
19
88
19
90
19
92
19
94
19
96
19
FDI North (exc. Belgium)
98
19
00
20
02
20
04
20
06
20
08
20
FDI South
33
Proposition 4: FDI composition is changing with p
Traded sector proportion of FDI
1.000
More FDI to
0.900
Traded sector
0.800
0.700
More FDI to
0.600
Non-Traded
0.500
0.400
0.300
0.200
0.100
0.000
0.650
0.700
0.750
0.800
0.850
0.900
0.950
1.000
1.050
1.100
1.150
1.200
relative price p
Traded-sector intensity of FDI:  
KT
1 

K  x
 (const)  H  p
K
1
  x
34
Fact 4: FDI in Manufacturing higher in the North
FDI in Manufacturing as %GDP
1
0.8
0.6
North
0.4
South
0.2
0
1991-98
1999-09
35
Fact 4: FDI in Manufacturing higher in the North
FDI in Real Estate as %GDP
1.4
1.2
1
0.8
0.6
0.4
0.2
0
North
South
1991-98
1999-09
36
Composition of FDI flows matters
• Potential advantages derived from FDI might differ markedly
across the primary, manufacturing and services sector
[UNCTAD (2001)].
• Ruane and Gorg (1999) showed that Ireland’s economy was
transformed because of the massive entrance of foreign
affiliates. MNEs were exporting over 70% of their output and
accounted for two thirds of manufacturing net output and half
of employment.
• The empirical discussions that concern spillover effects on
productivity and exports are often concentrated in the
manufacturing sector [see e.g. Aitken et al. (1997), Dimelis
and Louri (2002) and Greenway et al. (2004)].
37
Fact 4a: FDI in Manufacturing is positively
correlated with Trade Balances
FDI Manufacturing (% of
GDP)
1.600
-10.000
NL
1.400
IE
1.200
1.000
SP0.800
0.600
GR
FR
0.400
PT
0.200
-5.000
IT
FL
AT
DE
0.000
0.000
corr = 0.45
5.000
10.000
Trade balance (% of GDP)
38
Fact 4b: FDI in Real-estate is negatively correlated with
TRADE BALANCES
AT
1.800
1.600
PT
corr = -0.48
FDI Real Est. (%of
GDP)
1.400
1.200
SP 1.000
DE
FR
0.800
NLFL
0.600
0.400
0.200
-10.000 -8.000
-6.000
-4.000
0.000
-2.000 0.000
2.000
4.000
6.000
8.000
Trade balance (%of GDP)
Source: OECD and IMF
Belgium, Greece, Italy and Ireland not included, due to very few obs.
39
Financial Times
Germany’s unlikely optimists
By Ralph Atkins and James Wilson
Published: November
23 2010
… Says Mr Thomas Mayer, chief economist at Deutsche Bank,
On the underlying strength of Germany’s economic model:
“In the UK and US, we are now finding out that there was
too much expansion of the non-traded sector – too many
people in construction, too many people in real estate
and you could even say too many people in finance...
40
Proposition 5: External Balances diverge
If Traded sector FDI intensive: UP
If Non-traded sector FDI intensive: DOWN
Balance
North
B1
B0
South
B2
0
to
time
41
Fact 5: The evolution of Trade and CA Balances
10.00
8.00
6.00
4.00
2.00
20
08
20
06
20
04
20
02
20
00
19
98
19
96
19
94
19
92
19
90
19
88
19
86
19
84
19
80
-2.00
19
82
0.00
-4.00
-6.00
-8.00
EMU
-10.00
TB North
CANorth
TB South
CASouth
42
Why FDI has affected External Deficits ?
Recall Rybsczyncski Theorem (1955):
“An increase in a country’s endowment of a specific factor, will
increase the output of the sector that is intensive in that factor
and decrease the output in the other sector”.
Traded output:
(1  )  RK
  RH
YT 
K
H
   x
   x
(1  x)  RK
(  x)  RH
Non-traded output: YN  
K
H
(    x) p
(    x) p
43
N o r t h :   x    FD I intensive in the T raded sector
W ith a rise i n FD I:
O utput of the T raded sector w ill rise
O utput of the N on-T raded sector w ill shrink
T rade B a lance
T B  YT
- [ C  I ]
im pr oves
South:  >   x  FDI intensive in the Non-traded sector .
With a rise in FDI :
Output of the Traded sector will shrink
Output of the Non-Traded sector will rise
Trade Balance
TB  YT - [  C  I ]
deterior ates
44
Empirical Strategy
• We use the Pooled Mean Group estimator [Pesaran et al.
(1999)] where short-run coefficients, the speed of adjustment
and error variances are allowed to vary across countries, but the
long-run coefficients are constrained to be the same.
• An additional advantage of this model is that it yields consistent
estimates of the long-run parameters irrespectively of whether
the underlying regressors are stationary, nonstationary or
mutually cointegrated.
Yit  i (Yit 1  0i  1i X it )  11i X it   it
Where:
 i   (1   i ) ,  0i 
 10 i   11i
 20 i   21i

, 1i 
,  2i 
1  i
1  i
1  i
45
Empirical evidence for the relationship between
Spreads and Current account (quarterly data 1999-09)
(1)
(2)
-2.160***
(-3.328)
1.345***
(3.749)
-1.720**
(-2.888)
1.216***
(3.734)
-1.571**
(-2.821)
Long run coefficients
Current account
Government Debt
Fiscal balance
-
Short run coefficients
Speed of adjustment
Change in Current account
Change in Government Debt
Change in Fiscal balance
No. of obs.
Log likelihood
-0.198***
(-10.354)
-0.136
(-1.056)
1.510**
(2.731)
378
-1314.557
-0.223***
(-8.029)
0.308
(0.693)
1.459**
(2.693)
0.256*
(1.933)
378
-1289.381
46
Empirical evidence for the relationship between
Spreads and Current account
Using the Pooled Mean Group estimator we find evidence that
sovereign bond yield spreads are,
• Positively related to Government debt
• Positively related to fiscal deficits
• Positively related to external deficits 
See e.g. Alexopoulou et al. (2009), Attinasi et al. (2009), Barrios
et al. (2009)
47
Empirical evidence for the FDI effects on the trade
balance (annual data 1980-09)
North
(1)
Long run
coefficients
FDI
FDI manufacturing
FDI non-manufacturing
0.400***
(4.666)
-
South
(2)
2.099**
(2.282)
1.124**
(2.490)
(3)
-0.643**
(-2.593)
-
(4)
0.290
(0.379)
-0.608***
(-4.769)
Short run
coefficients
Speed of adjustment
-0.255**
(-3.523)
-0.116**
(-2.426)
-0.366***
(-3.953)
-0.507**
(-2.492)
No. of obs.
Log likelihood
145
-211.058
90
-121.549
144
-189.029
109
-126.644
48
Composition of FDI flows and trade balance
• The relationship of FDI with trade depends on the industry, country
and time period investigated [see Pontes (2004)]. 
Both export-oriented high tech manufacturing goods and labor
intensive textiles are included in manufacturing FDI, but they are
expected to affect a country’s productivity in a different way.
• The FDI effect can differ significantly between the short run and the
long run, depending on the nature of the imported goods in the short
run [see Fontagne (1999)]. 
Imported goods in the short run can be intermediate and capital
goods that serve as vehicles of positive externalities and improve
the host country’s trade balance in the long run though.
49
Trade balance determinants
1. We expect an increase in competitiveness ( Real Effective
Exchange Rate) to be positively related with external balance [see
e.g. Arghyrou and Chortareas (2008)]
2. According to Blanchard and Giavazzi (2002), increased economic
integration can result in large external deficits in transition
economies. Hence, we expect that an increase in the growth rate of
output to deteriorate the trade balance [see e.g. Abel and
Bernanke (2001), Gandolfo (2004)].
50
Evidence for the consistency of our results (NORTH)
(1)
(2)
annual data
quarterly
data
(3)
quarterly data
(1980-1998)
(4)
quarterly data
(1999-2009)
Long run
coefficients
FDI
0.485*** 0.089** 0.326** 0.061***
(6.578) (2.411) (2.374) (4.548)
-0.159***
(-3.386)
GROWTH -0.250**
(-2.017)
REER
-0.094*
(1.882)
-0.149
(-1.442)
-0.109**
(-2.411)
0.133
(0.982)
-0.083
(-1.237)
-0.171*
(-1.761)
Short run
coefficients
Speed of
adjustment
-0.254**
(-2.969)
-0.271***
(-3.749)
-0.348***
(-3.045)
-0.378**
(-2.250)
No. of obs.
Log likelihood
145
-177.089
523
-997.993
300
-547.833
215
-387.385
51
Evidence for the consistency of our results (SOUTH)
(1)
(2)
(3)
(4)
annual data
quarterly data
quarterly data
(1980-1998)
quarterly data
(1999-2009)
-1.014***
(-3.569)
-0.096***
(-3.879)
-0.419**
(-3.656)
-0.509***
(-3.754)
-0.149***
(-4.873)
-0.118
(-1.213)
1.553**
(4.339)
-0.176***
(-4.503)
-0.302**
(-2.050)
-0.346***
(4.454)
-0.160***
(-5.898)
-0.063
(-1.239)
Short run
coefficients
Speed of
adjustment
-0.424***
(-2.774)
-0.516*
(-1.929)
-0.305***
(-5.059)
-0.882**
(-3.618)
No. of obs.
Log likelihood
142
-155.928
503
-927.816
296
-538.379
203
-308.308
Long run
coefficients
FDI
REER
GROWTH
52
IN SEARCH OF A POLICY RE-FOCUSING
•
Focus on the New Asymmetries in EMU
•
The real issue: Upgrade productivity (how?)
•
Promote: Competitiveness, FDI to South EZ
•
Time for more policy coordination?
First step: Target Current Account Deficits
Second:
Coordinate wage-productivity agreements in EZ
Third:
Support productive investment esp. in the South
53