hagvoxtur2007regarding qualitythinking.ppt

Organisational entities include schools, charities clubs, sporting bodies, state owned enterprises, trust associations of persons and incorporated entities such as companies, through to multinational organizations such as United Nations, World Bank and International Monetary Fund.
1.7. In a democratic country, a government is the system through which powers are exercised and shared by the Legislature, Executive and the Judiciary with the goal of improving the quality of life for all citizens.

1. Three Lectures on Three Lectures on
Economic Efficiency Economic Efficiency
and Growthand Growth
Thorvaldur GylfasonThorvaldur Gylfason

2. Outline and aimsOutline and aims
Present a policy-oriented overview of Present a policy-oriented overview of
the theory and empirical the theory and empirical
evidence of economic growthevidence of economic growth
Trace linkages between economic Trace linkages between economic
growth and its main determinants: growth and its main determinants:
saving, investment, and economic saving, investment, and economic
efficiency efficiency
Exogenous vs. endogenous growthExogenous vs. endogenous growth
Liberalization, stabilization, Liberalization, stabilization,
privatizationprivatization
Education, institutions, natural Education, institutions, natural
resourcesresources

3. Outline and aimsOutline and aims
Lecture ILecture I
Saving, efficiency, and economic growthSaving, efficiency, and economic growth
Lecture IILecture II
Economic policy and growthEconomic policy and growth
Lecture IIILecture III
Education, natural resources, Education, natural resources,
institutions, and empirical evidenceinstitutions, and empirical evidence

4. IntroductionIntroduction
Growth theoryGrowth theory
As old as economics itselfAs old as economics itself
Smith, Marshall, Schumpeter, KeynesSmith, Marshall, Schumpeter, Keynes
Explicit growth theory started with Explicit growth theory started with
Harrod and Domar in 1940sHarrod and Domar in 1940s
Why important?Why important?
Unfashionable in 1960s and 1970sUnfashionable in 1960s and 1970s
Limits to growth, etc. Limits to growth, etc.
Growth and developmentGrowth and development
11

5. Growing apartGrowing apart
YearsYears
G
N
P

p
e
r

c
a
p
i
t
a
G
N
P

p
e
r

c
a
p
i
t
a
Country B: 2% a yearCountry B: 2% a year
Country A: 0.4% a yearCountry A: 0.4% a year
 InvestmentInvestment
 EfficiencyEfficiency
 InstitutionsInstitutions
 PolicyPolicy
Threefold Threefold
difference after difference after
60 years60 years
00 6060

6. Economic growth: Economic growth:
The short run vs. the long The short run vs. the long
runrun
TimeTime
N
a
t
i
o
n
a
l

e
c
o
n
o
m
i
c

o
u
t
p
u
t
N
a
t
i
o
n
a
l

e
c
o
n
o
m
i
c

o
u
t
p
u
t
Actual outputActual output
Potential outputPotential output
Business cyclesBusiness cycles
in the short runin the short run
Economic growthEconomic growth
in the long runin the long run
DownswingDownswing
UpswingUpswing

7. Ísland: Landsframleiðsla og Ísland: Landsframleiðsla og
framleiðslugeta 1945-2005framleiðslugeta 1945-2005
VerðbólguárinVerðbólguárin
: Yfir getu: Yfir getu
StöðnunStöðnun
Hófleg Hófleg
verðbólga: Undir verðbólga: Undir
getugetu

8. Other comparisonsOther comparisons
1)1)West-Germany vs. East-GermanyWest-Germany vs. East-Germany
2)2)Austria vs. Czech Republic Austria vs. Czech Republic
3)3)US vs. USSRUS vs. USSR
4)4)South Korea vs. North KoreaSouth Korea vs. North Korea
5)5)Taiwan vs. China Taiwan vs. China
6)6)Finland vs. EstoniaFinland vs. Estonia
See my See my Pictures of GrowthPictures of Growth
www.hi.is/~gylfason/pictures2.htmwww.hi.is/~gylfason/pictures2.htm
China vs. Europe:
China vs. Europe:
1:1 in 14001:1 in 1400
1:20 in 19891:20 in 1989

9. Further comparisonsFurther comparisons
1)1)Thailand vs. BurmaThailand vs. Burma
2)2)Mauritius vs. MadagascarMauritius vs. Madagascar
3)3)Botswana vs. NigeriaBotswana vs. Nigeria
4)4)Tunisia vs. MoroccoTunisia vs. Morocco
5)5)Spain vs. ArgentinaSpain vs. Argentina
6)6)Dominican Republic vs. HaitiDominican Republic vs. Haiti

10. Singapore and Malaysia:
GDP per capita 1965-2004
Constant
2000 US
dollars
(1.018)(1.018)
3939
= 2.0 = 2.0
4.0%4.0%
5.8%5.8%

11. Botswana and Nigeria:
GNP per capita 1965-2004
Constant
2000 US
dollars
(1.065)(1.065)
3939
= 11.7 = 11.7
0.6%0.6%
7.1%7.1%

12. Spain and Argentina :
GNP per capita 1965-2004
(1.019)(1.019)
3939
= 2.1 = 2.1
0.6%0.6%
2.7%2.7%

13. Mauritius and
Madagascar: GNP per
capita 1965-2004
Constant
2000 US
dollars
(1.055)(1.055)
3939
= 8.1 = 8.1
-1.2%-1.2%
4.3%4.3%

14. Ireland and Greece:
GNP per capita 1965-2004
Constant
2000 US
dollars
(1.016)(1.016)
3939
= 1.9 = 1.9
2.6%2.6%
4.2%4.2%

15. Basic growth theoryBasic growth theory
A.A.Harrod-Domar modelHarrod-Domar model
B.B.Solow modelSolow model
C.C.Endogenous growth modelEndogenous growth model
Let’s do the algebra
Let’s do the algebra

16. Harrod-Domar modelHarrod-Domar model
sYS
vYK
IS
KKIsYS 
vYYvsY 
  YvYvs 


v
s
Y
Y

v
s
g
Two assumptionsTwo assumptions
Implications for growthImplications for growth 03.004.0
3
21.0
g
grossgrossnetnetreplacementreplacement
Fixed saving rate
Fixed saving rate
Fixed capital/output ratio
Fixed capital/output ratio

17. Harrod-Domar modelHarrod-Domar model

v
s
g
Three propositions about growthThree propositions about growth
0
ds
dg
0
dv
dg
0
d
dg
Saving increases growth
Saving increases growth
Efficiency increases growth
Efficiency increases growth
Depreciation reduces growth
Depreciation reduces growth
v = K/Yv = K/Y
1/v = Y/K1/v = Y/K

18. Stöldrum við:Stöldrum við:

g = sE - g = sE - , svo , svo s = 0 s = 0 =>=> g < 0 g < 0
Það þarf lágmarkssparnað til að vaxa til að Það þarf lágmarkssparnað til að vaxa til að
vega á móti afskriftumvega á móti afskriftum
Sum lönd eru svo fátæk, að þau komast Sum lönd eru svo fátæk, að þau komast
ekki upp á þröskuldinn og eru því dæmd ekki upp á þröskuldinn og eru því dæmd
til neikvæðs hagvaxtar: til neikvæðs hagvaxtar: fátæktargildrafátæktargildra
Er hægt að hjálpa þeim upp á þröskuldinn? Er hægt að hjálpa þeim upp á þröskuldinn?

Einfalt líkan um Einfalt líkan um
innri hagvöxtinnri hagvöxt

19.    
                                       
 
                                       
 
                                              
 
                                              
Fátækt um heiminn Fátækt um heiminn 20012001
0
200
400
600
800
1000
1200
Fólk sem lifir á minna en
dollara á dag
Fólk sem lifir á minna en
tveim dollurum á dag
Austur-Asía
Suður-Asía
Afríka
MilljónirMilljónir
Samtals 1,1 Samtals 1,1
milljarður, var 1,5 milljarður, var 1,5
milljarðar 1981 milljarðar 1981
Samtals 2,8 milljarðarSamtals 2,8 milljarðar

20.    
                                       
 
                                       
 
                                              
 
                                              
16
50
31
77
45
75
0
10
20
30
40
50
60
70
80
90
Fólk sem lifir á minna en
dollara á dag
Fólk sem lifir á minna en
tveim dollurum á dag
Austur-Asía
Suður-Asía
Afríka
% af mannfjölda% af mannfjölda
Fátækt um heiminn Fátækt um heiminn 20012001

21. Er hægt að lyfta Er hægt að lyfta öllumöllum upp fyrir dollara á upp fyrir dollara á
dag?dag?
Hvað myndi það kosta? – að útrýma sárri Hvað myndi það kosta? – að útrýma sárri
fátæktfátækt
Fjöldi fólks undir dollara á dag: 1,1 Fjöldi fólks undir dollara á dag: 1,1
milljarðurmilljarður
Meðaltekjur þeirra eru 77 sent á dag, þurfa Meðaltekjur þeirra eru 77 sent á dag, þurfa
1,08 dollara1,08 dollara
Munurinn er 31 sent á dag, eða 113 dollarar á Munurinn er 31 sent á dag, eða 113 dollarar á
áriári
Heildarkostnaðurinn er því 124 milljarðar Heildarkostnaðurinn er því 124 milljarðar
dollara á ári, eða 0,6% af VLF í iðnríkjumdollara á ári, eða 0,6% af VLF í iðnríkjum
Minna en þau hafa lofað! – og ekki efnt Minna en þau hafa lofað! – og ekki efnt
Er hægt að útrýma Er hægt að útrýma
fátækt?fátækt?

22. Solow modelSolow model
Four assumptionsFour assumptions
Full employment
Full employment
aa
KALY


1
)(
nt
eLL
0

F
YY
KKIsYS 
Constant growth of labor force
Constant growth of labor force
Constant returns to scale
Constant returns to scale
gaan
K
Y
saang )1()1( 






 
Saving equals investment
Saving equals investment
ng
.constA
Output growth equals labor force growth
Output growth equals labor force growth

23. Solow modelSolow model
Endogenous output/capital ratioEndogenous output/capital ratio






 
k
y
saang )1(
ExogenousExogenous
EndogenousEndogenous
ExogenousExogenous
aa
KALY


1
)(
a
Aky


1
a
Ak
k
y


Need to determine k and hence y/k
Need to determine k and hence y/k

24. Solow modelSolow model
Dynamic stability of output/capital ratioDynamic stability of output/capital ratio
L
K
k
LKk


nsAkn
k
y
sLKk
a




0
)1(

a
asAk
dk
kd

Stable equilibrium
Stable equilibrium
s
n
k
y
k

0


25. Solow modelSolow model
Dynamic stability of output/capital ratioDynamic stability of output/capital ratio
0
)1(

a
asAk
dk
kd

Stable equilibrium
Stable equilibrium
k
k

EE

26. Solow modelSolow model
Two equations in two unknowns, y and kTwo equations in two unknowns, y and k
Long-run equilibrium
Long-run equilibrium
a
Aky


1
k
s
n
y 







a
n
sA
k
1









a
a
a
A
n
s
y
1
1










27. O
u
t
p
u
t

p
e
r

h
e
a
d
Capital per worker
Output/capital ratio
a
Aky


1
k
s
n
y 







Solow modelSolow model
EE
Comparative statics:
Comparative statics:
E moves in response to
E moves in response to
changes in s, A, n, and
changes in s, A, n, and 

28. Solow modelSolow model
Four propositions about long-run growthFour propositions about long-run growth
a
a
a
A
n
s
y
1
1









0
ds
dy
0
dA
dy
0
dn
dy
0
d
dy
Increased saving increases income per capita
Increased saving increases income per capita
Increased efficiency increases income per capita
Increased efficiency increases income per capita
Increased population growth reduces income per capita
Increased population growth reduces income per capita
Increased depreciation reduces income per capita
Increased depreciation reduces income per capita

29. Solow modelSolow model
Now allow technological progressNow allow technological progress
qt
eAA
0
aa
KALY


1
)(
gaqna
K
Y
saqnag )1()()1()( 






 
qng
Technological progress at a fixed rate
Technological progress at a fixed rate
Constant returns to scale
Constant returns to scale
Growth depends solely
Growth depends solely
on technological
on technological
progressprogress
GDP grows at same rate as
GDP grows at same rate as
population plus technology,
population plus technology,
so GDP per capita grows at
so GDP per capita grows at
rate rate qq

30. Closed-form solution to Closed-form solution to
Solow modelSolow model
aa
KALY


1
)(
nt
eLL
0
qt
eAA
0
tqn
eLAAL
)(
00


ALYy/ˆ
ALKk/
ˆ

KsYKIK  
kys
AL
K
ˆ
ˆ

a
ky


1ˆ
ˆ
Labor-augmenting
Labor-augmenting
technological progress
technological progress

31. Closed-form solution to Closed-form solution to
Solow modelSolow model
kqknkysAL
LA
K
AL
K
AL
K
k
ˆˆˆ
ˆ)(
ˆ
22














 
qnksqn
k
y
s
k
k
g
a
k











ˆ
ˆ
ˆ
ˆ
ˆ
ˆ
  skqn
k
k
a








ˆ
ˆ
ˆ


 skqnkk
aa

 ˆˆˆ
1

a
qn
s
k
1
ˆ












32. Closed-form solution to Closed-form solution to
Solow modelSolow model
a
kv
ˆ











k
k
a
v
v
ˆ
ˆ
v
k
k
a
v











ˆ
ˆ
 svqnv
a
kv
a






 
1
ˆ
  tqnaaa
e
qn
s
k
qn
s
kv





















0
ˆˆ
a(a(+n+q) is the speed of convergence
+n+q) is the speed of convergence

33. Closed-form solution to Closed-form solution to
Solow modelSolow model
a
t qn
s
k
1
ˆ
lim










 
a
a
t qn
s
y

 









1
ˆlim

a
a
qt
a
a
tt qn
s
eA
qn
s
A
L
Y
y

 

























1
0
1
limlim

timetime
k
ˆ
0
ˆ
k

34. Solow model: Solow model:
ConvergenceConvergence
175.0
04.0


te
t
408.0
04.0


te
t
 kqnak
ˆˆ

  kkkk
ˆ
04.006.07.0
ˆ
04.001.001.07.0
ˆ

Takes 17 years to close half the gap
Takes 17 years to close half the gap
Takes 40 years to close 80% of the gap
Takes 40 years to close 80% of the gap

35. Solow model: Solow model:
ConvergenceConvergence
O
u
t
p
u
t

p
e
r

h
e
a
d
Capital per worker
a
Aky


1
k
s
n
y 







Rich country’s Rich country’s
initial income per headinitial income per head
Poor country’s Poor country’s
initial income per headinitial income per head
Poor country must grow
Poor country must grow
faster if it is to catch up
faster if it is to catch up

36. An Increase in the An Increase in the
Saving RateSaving Rate
O
u
t
p
u
t

p
e
r

h
e
a
d
O
u
t
p
u
t

p
e
r

h
e
a
d
Capital per workerCapital per worker
E
F
C’
C
P

37. O
u
t
p
u
t

p
e
r

h
e
a
d
Capital per workerCapital per worker
E
F
P
P’
C
An Increase in Static An Increase in Static
EfficiencyEfficiency

38. O
u
t
p
u
t

p
e
r

h
e
a
d
O
u
t
p
u
t

p
e
r

h
e
a
d
Capital per workerCapital per worker
F
E
C
C’
P
An Increase in Population An Increase in Population
Growth or DepreciationGrowth or Depreciation

39. O
u
t
p
u
t

p
e
r

h
e
a
d
O
u
t
p
u
t

p
e
r

h
e
a
d
Capital per workerCapital per worker
F
E
C
C’
P
P’
An Increase in Dynamic An Increase in Dynamic
Efficiency (Technical Progress)Efficiency (Technical Progress)

40. Solow model: Solow model:
ConclusionConclusion
Three main points to noteThree main points to note
Long-run growth is exogenous: Long-run growth is exogenous: g = n + qg = n + q
No role for economic forces, policy or No role for economic forces, policy or
institutions, just technologyinstitutions, just technology
But education is good for growthBut education is good for growth
Model implies convergenceModel implies convergence
Poor countries grow more rapidly than Poor countries grow more rapidly than
richrich
The medium term can be quite longThe medium term can be quite long
Growth is endogenous for a long whileGrowth is endogenous for a long while

41. Solow model with Solow model with
educationeducation
aa
KAHY


1
)(
bt
LeH
bqng 
H = skilled labor
H = skilled labor
L = raw labor, b = years of schooling
L = raw labor, b = years of schooling
AH grows at n + q + bAH grows at n + q + b
Education stimulates long-run growthEducation stimulates long-run growth
0
ds
dg

42. Endogenous growthEndogenous growth
aa
KALY


1
a
LKEA )/(
qng
E = efficiencyE = efficiency
EKY
sEg
E = 1/v in Harrod-Domar
E = 1/v in Harrod-Domar
Harrod-Domar, again, without the drawbacks
Harrod-Domar, again, without the drawbacks
Two equations in two unknowns, g and q
Two equations in two unknowns, g and q

43. Endogenous growthEndogenous growth
Technological progress, q
E
c
o
n
o
m
ic

g
r
o
w
t
h
,

g
45°
Population growth
O
AC
Saving rate
times efficiency
minus depreciation
B
sEg
qng
0
ds
dg
0
dE
dg
Let’s take four examples
Let’s take four examples

44. Beint eða bogið?Beint eða bogið?
aa
KALY


1 a
L
K
A
L
Y








1
Y/LY/L
K/LK/L
a < 1a < 1 þýðir þýðir
jákvæð en jákvæð en
minnkandi minnkandi
markaframleiðni
markaframleiðni
fjármagnsfjármagns
Cobb-Douglas framleiðslufall
Cobb-Douglas framleiðslufall

45. Beint eða bogið?Beint eða bogið?
aa
KALY


1 a
L
K
A
L
Y








1
Y/LY/L
K/LK/L
Hækkun Hækkun ss þýðir þýðir
hækkun hækkun K/LK/L úr B í C úr B í C
og þar með lækkun
og þar með lækkun
Y/K, Y/K, svo aðsvo að sY/K sY/K
hækkar í bráð, en
hækkar í bráð, en
ekki til langframa
ekki til langframa
Cobb-Douglas framleiðslufall
Cobb-Douglas framleiðslufall
BB
CC

46. Beint eða bogið?Beint eða bogið?
Hagkvæmni fer eftir fjármagni á mann
Hagkvæmni fer eftir fjármagni á mann
aa
KALY


1 a
L
K
A
L
Y








1
a
L
K
EA 






L
K
E
L
Y
 EKY

47. Beint eða bogið? Beint eða bogið?
Framleiðslufallið er
Framleiðslufallið er bein línabein lína: :
Föst markaframleiðni fjármagns
Föst markaframleiðni fjármagns
aa
KALY


1 a
L
K
A
L
Y








1
a
L
K
EA 






L
K
E
L
Y
 EKY
YY
KK
11
EE
E = Y/K er fasti
E = Y/K er fasti

48. A tax on education and A tax on education and
endogenous growthendogenous growth
G = government spending on education
G = government spending on education
G is financed by tax on capital
G is financed by tax on capital
Constant returns to capital
Constant returns to capital
AGKY
tKG
EKKAtY 
AtE
Atsg
A tax to finance education
A tax to finance education
is good for growth
is good for growth
11

49. Inflation, money, and Inflation, money, and
endogenous growthendogenous growth
Output is made by financial and real capital
Output is made by financial and real capital
K
P
M
Y 






K
PK
M
Y 





 






PK
M
E


1
1
PK
M










1
1
sg
Inflation reduces the use of financial capital
Inflation reduces the use of financial capital
22
Inflation impedes growth
Inflation impedes growth

50. Education and endogenous Education and endogenous
growth, againgrowth, again
R&D model (Romer)R&D model (Romer)
LbAY )1(
BbLAA
BbL
A
A


nBbLg 
b = fraction of labor force engaged in R&D
b = fraction of labor force engaged in R&D
bL = number of workers engaged in R&D
bL = number of workers engaged in R&D
A = stock of existing knowledge
A = stock of existing knowledge
Growth depends on R&D
Growth depends on R&D
0
ds
dg
33
0
db
dg

51. Education and endogenous Education and endogenous
growth, once moregrowth, once more
Less education
Less education
means less growth
means less growth
Human capital model (Lucas)Human capital model (Lucas)
aa
KALY


1
a
cA
aa
KcLY


1
)(
h
c
c


1
c = human capital
c = human capital
h = fraction of time spent on work
h = fraction of time spent on work
hng 1
0
ds
dg
Let’s look at some evidence, but first …
Let’s look at some evidence, but first …
44

52. How growth becomes How growth becomes
endogenousendogenous
Solow model: when s rises, E = Y/K falls due Solow model: when s rises, E = Y/K falls due
to decreasing returns to capital, so g stays to decreasing returns to capital, so g stays
putput
Endogenous growth: when s rises, E stays Endogenous growth: when s rises, E stays
put due to constant returns to capital, so g put due to constant returns to capital, so g
risesrises
aa
KALY


1
)(
a
Aky


1
a
Ak
k
y


EKY )(kfE
k
y
K
Y


53. Empirical growth researchEmpirical growth research
1)1)Cross-country regressionsCross-country regressions
Large samples, beginning in 1960 or Large samples, beginning in 1960 or
19701970
2)2)Cross sections vs. panelsCross sections vs. panels
3)3)Averages vs. initial values of Averages vs. initial values of
independent variablesindependent variables
Cost of simultaneity bias vs. cost of Cost of simultaneity bias vs. cost of
discarding available datadiscarding available data
4)4)Recursive modeling vs. instrumentsRecursive modeling vs. instruments
5)5)Levels of income vs. rates of growthLevels of income vs. rates of growth

54. Recursive modelingRecursive modeling
Growth regressionGrowth regression
g = g = aa
00 – – aa
11yy
00 + + aa
22x + x + aa
33zz (1)(1)
where x is exogenous and z is endogenouswhere x is exogenous and z is endogenous
z = z = bb
00 + + bb
11yy
00 – – bb
22xx (2)(2)
where z is, say, education and x is natural where z is, say, education and x is natural
resource relianceresource reliance
Eq. (2) makes z exogenous, so (1) and (2) Eq. (2) makes z exogenous, so (1) and (2)
can be estimated by OLScan be estimated by OLS
TSLS calls for instruments that help explain z TSLS calls for instruments that help explain z
without being correlated with g: Not without being correlated with g: Not
easyeasy

55. Levels of income vs. Levels of income vs.
rates of growthrates of growth
cxbyag 
0

56. Levels of income vs. Levels of income vs.
rates of growthrates of growth
cxbyag 
0
)
100
()
100
1ln(
00
g
Ty
g
Tyy 

57. Levels of income vs. Levels of income vs.
rates of growthrates of growth
cxbyag 
0
)
100
()
100
1ln(
00
g
Ty
g
Tyy 
)(
100
00 cxbya
T
yy 

58. Levels of income vs. Levels of income vs.
rates of growthrates of growth
cxbyag 
0
)
100
()
100
1ln(
00
g
Ty
g
Tyy 
)(
100
00 cxbya
T
yy 
xyy  
0
100
T
a
100
T
c

59. Levels of income vs. Levels of income vs.
rates of growthrates of growth
cxbyag 
0
)
100
()
100
1ln(
00
g
Ty
g
Tyy 
)(
100
00 cxbya
T
yy 
xyy  
0
100
T
a
100
T
c
100
1
T
b
Conditional Conditional
convergence requires b
convergence requires b
> 0 > 0  < 1< 1
One-to-one One-to-one
correspondence between
correspondence between
parametersparameters

60. Absolute convergence: Absolute convergence:
Growth ratesGrowth rates
y = -0,864x + 8,305
R² = 0,182
-10
-5
0
5
10
15
4 6 8 10 12
G
r
o
w
t
h

o
f

G
D
P

p
e
r

c
a
p
i
t
a

1
9
6
0
-
2
0
0
0

(
%
)
Log of GNP per capita 1960
b = 0.864b = 0.864
165 countries
Equatorial
Guinea

61. Absolute convergence: Absolute convergence:
Levels of incomeLevels of income
y = 0,657x + 3,282
R² = 0,452
4
6
8
10
12
4 5 6 7 8 9 101112
L
o
g
o
f
G
N
P

p
e
r

c
a
p
i
t
a
2
0
0
0
Log of GNP per capita 1960
4
45º
 = 0.658= 0.658
= 1 – 0.4= 1 – 0.40.8640.864
Conditional convergence
Conditional convergence
is stronger, as we will
is stronger, as we will
seesee
165 countries
Equatorial
Guinea

62. From efficiency to growthFrom efficiency to growth
Basic resultBasic result
If it – anything! – increases economic If it – anything! – increases economic
efficiency, it is also good for growthefficiency, it is also good for growth
Follows from Harrod-Domar model as Follows from Harrod-Domar model as
well as from endogenous-growth well as from endogenous-growth
theory and also, as a proposition theory and also, as a proposition
about the medium run, from the about the medium run, from the
Solow modelSolow model
In practice, Solow model and In practice, Solow model and
endogenous growth are hard to endogenous growth are hard to
distinguishdistinguish
So, let’s look more closely at efficiency So, let’s look more closely at efficiency

63. Liberalization Liberalization Increases Increases
Economic EfficiencyEconomic Efficiency
GG
EE
CC
Traditional Traditional
outputoutput
Modern outputModern output
HH
DD Domestic, Domestic,
distorted distorted
price ratioprice ratio
OO

64. Liberalization Liberalization Increases Increases
Economic EfficiencyEconomic Efficiency
GG
FF
AA BCC
Output Output
gaingain
Traditional Traditional
outputoutput
Modern outputModern output
DD
HH
EE
Price Price
distortiondistortion
World price ratioWorld price ratio
Domestic, Domestic,
distorted distorted
price ratioprice ratio
If output gain = E and
price distortion = c,
then E = mc
2

OC = modern output
CA = traditional output
OA = total output
OO

65. Liberalization Liberalization Increases Increases
Economic EfficiencyEconomic Efficiency
GG
FF
AA BCC
Output Output
gaingain
Traditional Traditional
outputoutput
Modern outputModern output
DD
HH
EE
Price Price
distortiondistortion
World price ratioWorld price ratio
Domestic, Domestic,
distorted distorted
price ratioprice ratio
OO
JJ KK
ImportsImports
ExportsExports
Welfare Welfare
gaingain

66. Liberalization Liberalization Increases Increases
Economic EfficiencyEconomic Efficiency
GG
FF
AA BCC
Traditional Traditional
outputoutput
Modern outputModern output
DD
HH
EE
Price Price
distortiondistortion
OO
JJ
Welfare Welfare
gaingain
MM
NN
QQ
Transition
takes time:
From E to F
via M, N,
and Q

67. Liberalization Liberalization Increases Increases
Economic EfficiencyEconomic Efficiency
DD
GG
FF
EE
AA
BB
Price Price
distortiondistortion
JJ
OO
HH
World price World price
ratioratio
CC
AB = static gainAB = static gain
BC = dynamic gainBC = dynamic gain
AC = AB + BC = total gainAC = AB + BC = total gain
Productivity Productivity
gaingain
KK
WelfareWelfare
gaingain
Traditional Traditional
outputoutput
Modern outputModern output

68. Stabilization Stabilization Increases Increases
Economic EfficiencyEconomic Efficiency
Financial capitalFinancial capital
Real Real
capitalcapital
Undistorted Undistorted
price ratioprice ratio
GG
FF
EE
AA
InflationInflation
distortiondistortion
45°
Output afterOutput after
Output beforeOutput before
Distorted Distorted
price ratioprice ratio
CC
DD
HH
OO BB
Output Output
gaingain
If output gain = E and
inflation distortion = c,
then, again, E = mc
2

69. Privatization Privatization Increases Increases
Economic EfficiencyEconomic Efficiency
Private outputPrivate output
Public Public
outputoutput
Undistorted Undistorted
price ratioprice ratio
D
G
Distorted Distorted
price ratioprice ratio
F
E
A BN
Price andPrice and
qualityquality
distortiondistortion
H
O
J
K
45°
Output Output
gaingain

70. From efficiency to growth: From efficiency to growth:
Same story time and again Same story time and again
Free tradeFree trade is good for growth is good for growth
–Reduces the inefficiency that results from Reduces the inefficiency that results from
restrictions on trade restrictions on trade
Price stabilityPrice stability is good for growth is good for growth
–Reduces inefficiency resulting from inflationReduces inefficiency resulting from inflation
PrivatizationPrivatization is good for growth is good for growth
–Reduces inefficiency resulting from SOEsReduces inefficiency resulting from SOEs
EducationEducation is good for growth is good for growth
–Reduces the inefficiency that results from Reduces the inefficiency that results from
inadequate educationinadequate education

71. Investment and growth, Investment and growth,
1960-20001960-2000
r = rank
correlation
164 countries164 countries

72. Exports and growth, Exports and growth,
1960-20001960-2000
r = rank
correlation
163 countries163 countries

73. Education and growth, Education and growth,
1960-20001960-2000
131 countries131 countries
A 25 point increase in secondary-
school enrolment goes along
with an increase in per capita
growth by 1% per year
Diminishing returns: The
additional benefit from education
becomes smaller as enrolment
increases

74. Life expectancy and Life expectancy and
growth, 1960-2000growth, 1960-2000
147 countries147 countries

75. Corruption and growth, Corruption and growth,
1960-20001960-2000
88 countries88 countries

76. Inflation and growth, Inflation and growth,
1960-20001960-2000
164 lönd164 lönd
40 ár40 ár
Hagvöxtur stendur í
Hagvöxtur stendur í
öfugu sambandi við
öfugu sambandi við
verðbólgu milli landa
verðbólgu milli landa
– og sést með berum
– og sést með berum
augumaugum
Verðbólgubjögun Verðbólgubjögun
= = /(1+/(1+ ) )

77. Political liberties and Political liberties and
growth, 1960-2000growth, 1960-2000
65 countries65 countries

78. Democracy and growth, Democracy and growth,
1960-20001960-2000
144 countries144 countriesr = 0.48

79. Nonmanufacturing output Nonmanufacturing output
and growth, 1960-2000and growth, 1960-2000
164 countries164 countries
Now, let’s Now, let’s
run some run some
regressionregression
ss
r = -0.64

80. Growth regressionsGrowth regressions
Based on World Bank dataBased on World Bank data
World Development Indicators, World Development Indicators,
published each year on CDpublished each year on CD
Wide coverage: 208 countries, 42 yearsWide coverage: 208 countries, 42 years
Could also use Penn data (compiled by Could also use Penn data (compiled by
Summers and Heston), but they Summers and Heston), but they
cover fewer countriescover fewer countries
Here, we report cross-sectional evidence, Here, we report cross-sectional evidence,
representing each country by a representing each country by a
single observation for each variablesingle observation for each variable

81. Keep an eye on the number of observations
Keep an eye on the number of observations
INITIAL is the log of per capita GNP
INITIAL is the log of per capita GNP
in 1960, so the coefficient describes
in 1960, so the coefficient describes
the speed of convergence, nearly
the speed of convergence, nearly
1% 1%

82. INVEST is the ratio of
INVEST is the ratio of
investment to GDP
investment to GDP
An increase in investment by 8% of
An increase in investment by 8% of
GDP increases growth by 1 percentage
GDP increases growth by 1 percentage
pointpoint

83. LOGENROL is the log of
LOGENROL is the log of
secondary-school enrolment
secondary-school enrolment
(net)(net)
An increase in secondary-school enrolment by 60% (e.g.,
An increase in secondary-school enrolment by 60% (e.g.,
from 50% to 80%) increases growth by 1 percentage point
from 50% to 80%) increases growth by 1 percentage point

84. PRIMGDP is the share of
PRIMGDP is the share of
primary production in GDP
primary production in GDP
An increase in the share of primary production in
An increase in the share of primary production in
GDP by 25% increases growth by 1 percentage
GDP by 25% increases growth by 1 percentage
pointpoint

85. DISTORTDISTORT = = /(1+/(1+ ) )
A decrease in inflation from 40% to zero increases
A decrease in inflation from 40% to zero increases
growth by nearly 1 percentage point
growth by nearly 1 percentage point

86. POPGRO is the rate of growth
POPGRO is the rate of growth
of populationof population

87. DUMMYAFR is a dummy for
DUMMYAFR is a dummy for
sub-Saharan Africa
sub-Saharan Africa

88. Conditional Conditional
convergenceconvergence
, as before, as before
Dependent variable is now the
Dependent variable is now the
level of per capita GNP: same story
level of per capita GNP: same story

89. ConclusionConclusion
The End
The End
Saving and efficiency Saving and efficiency
are good for growthare good for growth
Efficiency gains take Efficiency gains take
many different forms many different forms
Liberalization, stabilization, privatizationLiberalization, stabilization, privatization
Conversion of inputs into output is not Conversion of inputs into output is not
solely a matter of technology, but also solely a matter of technology, but also
efficiency, so efficiency, so economic policy matterseconomic policy matters

90. Classroom discussion
Classroom discussion