Transcript Inclusive Growth Framework

Income Benchmark
Applied Inclusive Growth Analytics Course
June 29, 2009
Leonardo Garrido
Presentation plan






Establish a rationale for benchmarking
Define benchmark guidelines
Identify elements of an income benchmark
Identify elements of a potential growth
benchmark
Show practical issues on Income
Benchmarking
Tool demo
2
Benchmarking, why?

Countries always looking up at peers / role models

Theoretical foundations: Subjective well-being theory (Happiness
Economics)


Satisfaction from material and non material goods
Interdependent utility functions


For better assessment of a country growth performance and potential


As an anchor.
Is the country catching up?




http://www.brookings.edu/views/articles/graham/200509.pdf
(income benchmark tool example)
Will it be able to achieve a given target?
To help identify shadow prices in a Growth Diagnostic exercise
Because, despite heterogeneity and idiosyncratic elements, empirical
regularities have been found in cross country analysis

Predictions from neoclassical growth models : Conditional Convergence

Growth forecasting tool http://go.worldbank.org/RQUIVOP9X0
3
Benchmark guidelines (I)

The choice of relevant benchmark countries depends on the
question one would like to be answered

Are we interested in expected effect of proposed policies?



Is it about potential or expected growth outcomes?


Role models
Is it about linkages?


Look up at similar countries and initial conditions
Is it about achieving long run targets?


Look up at “similar” countries
Ideally, one should aspire to a “natural experiment” where the selected
benchmark is a replica in all but one respect to the country under study
Correlate countries on variables of interest
Is it about outcomes?

Empirical identification of relevant groups
4
Benchmark guidelines (II)

Similar countries to compare with


“Similarity”: Attribute that is thought to primarily determine a given outcome
Candidates for similarities include: Initial conditions, income per capita,
geographic characteristics, productive structure, export patterns…


Role models


Countries which performance or welfare indicator wants to be attained.
Correlate countries


Collier & O’Connell, Opportunities and Choices: http://users.ox.ac.uk/~econpco/research/africa.htm
Those affecting /affected by analyzed countries by means of some
transmission mechanism.
Outcomes




Growth Commission, “13 Success Stories” http://www.growthcommission.org/index.php
Growth Accelerations, Hausmann et al (2004) http://www.nber.org/papers/w10566
Growth Patterns, Pritchett (2000) http://papers.ssrn.com/sol3/papers.cfm?abstract_id=629200
Sustained Growth in Africa, Johnson et al. (2007) http://www.imf.org/external/pubs/ft/wp/2007/wp0752.pdf
5
Elements of an Income Benchmark (I)

Basically, one aims to answer some of the following questions:






Notice that our questions refer to Income Levels. We are ignoring
issues regarding the growth rate, poverty and income distribution.



How long would it take to analyzed country to reach benchmark’s current
income level at a given (hypothetical) growth rate?
How long would it take to analyzed country to catch up with benchmark
country at hypothetical growth rate?
How long would it take to attain a given income level category?
What growth rate would it be necessary to catch up / reach current
income level of benchmark country or reach threshold of next income
level category?
What income level will it be obtained growing at hypothetical growth
rate during specified period of time?
Growth potential to be addressed in next session (Growth
Decomposition)
Poverty-growth links addressed tomorrow (Kenneth Simler)
Also ignores the relationship between the cycle and the trend
6
Elements of an Income Benchmark:
A reality check

Is the hypothetical / desired income level feasible?

Feasibility: Country is (and will not be) resource constrained

Input availability / Dynamic input formation compatible with income
target






Observe we are touching upon the growth rate
Compare desired with historical trends
Scenarios for TFP: Linked to: Reform, Composition and level of government
expenditures…
External – Internal Balance: Are Saving, Investment and Current
Account Balances path compatibles with a given income target /
growth rate?
Political economy issues: Will the country be able to do
what it takes to accelerate growth / move to a higher income
level?
What are the long term growth implications of short term
fluctuations? (shocks)
7
Potential growth benchmark

Normally interested in “accelerating” and “sustaining”
“high” growth rates


Potential growth: Growth rate a country can achieve and
sustain at country’s natural rate of unemployment of
productive factors (natural rate of output growth).


These are empirical definitions. See references above (p.5)
More on this in next session
From benchmarking point of view, it is reasonable to
compare “initial conditions” in analyzed countries vis a vis
initial conditions present benchmark countries when growth
began to accelerate.

(Johnson et al, 2007)
8
Income Benchmark Tool Nuts and Bolts (I)
Measuring GDP growth

2 ways of calculating annual growth rate of GDP for period T>2:
  yT
 ln 
  y0

 T



 






From initial vs final period GDP

From OLS regression using all data

Methods yields slightly different results, but are highly correlated

Excel example
y  e
1
ln  y       @ trend
9
Income Benchmark Tool Nuts and Bolts (II)
Formulas


Basic formula
Yt  Y0  e g t
How long would it take to analyzed country to reach
ln Yt   ln Y0   g  t 
benchmark’s income level, at analyzed country’s
hypothetical growth rate (g)?
ln Yt Y0
– Y0 is income level of analyzed country
t
– Yt is income level of benchmark country
g
– Solve for unknown t



Time to catch up if analyzed country grows at (ga)
and benchmark country grows at gb?
– Y0 is income level of analyzed country
– Yt is income level of benchmark country
– Solve for unknown t
– How to interpret negative result?

Yt  e g b t  Y0  e g a t 
ln Yt   gb  t  ln Y0   g a  t 
t
ln Yt Y0 
g a  g b 
Solve for “g” instead of “t” to answer questions regarding the growth rate
10
Income Benchmark Tool Nuts and Bolts (III)
Excel formulas

Excel commands to calculate historical growth rates:
Year ->
Ypc_analyzed
Ypc_benchmark
1
100.0
1,000.0
2
105.0
1,010.0
3
110.3
1,020.1
4
115.8
1,030.3
5
121.6
1,040.6
growth rate
4.88%
1.00%
4.6
6.9
4.7
6.9
4.7
6.9
4.8
6.9
4.8
6.9
4.88%
1.00%
ln(Ypc_analyzed)
ln(Ypc_benchmark)
9/A
10
11
12
13
14
15
Year ->
Ypc_analyzed
Ypc_benchmark
B
1
100
1000
C
2
105
1010
D
3
110.25
1020.1
E
4
115.7625
1030.301
F
5
121.550625
1040.60401
G
H
I
growth rate
=LINEST(LN(C11:G11))
=LINEST(LN(C12:G12))
ln(Ypc_analyzed)
ln(Ypc_benchmark)
=LN(C11)
=LN(C12)
=LN(D11)
=LN(D12)
=LN(E11)
=LN(E12)
=LN(F11)
=LN(F12)
=LN(G11)
=LN(G12)
=SLOPE(C14:G14,C10:G10)
=SLOPE(C15:G15,C10:G10)
11
Income Benchmark Tool Nuts and Bolts (IV).
GDP units




Real Prices vs Current Prices
Local Currency Units vs Foreign Currency Units
Parity Purchasing Power
Alternative Methods: World Bank Atlas
12
GDP comparisons
Bottom poorest countries, in per capita GNI
(Current US$ vs Current US$ at PPP) 2005
GDP Per Capita: Current US$ vs PPP US$
12
LN (GDPpc US$)
11
10
9
8
7
6
5
4
4
6
8
10
12
LN(GDPpc PPP US$)
7.5
Zambia: GDP per capita. Real US$ vs Real PPP
6.5
6
US$
90
120
120
160
170
180
220
220
250
250
270
270
290
290
300
330
340
340
350
400
GNI per capita, PPP (Current $)
Ranking Country
2
Congo, Dem. Rep.
3
Liberia
1
Burundi
6
Guinea-Bissau
5
Eritrea
7
Sierra Leone
9
Niger
4
Ethiopia
13
Mozambique
8
Malawi
17
Central African Republic
18
Togo
10
Rwanda
11
Uganda
14
Madagascar
15
Nepal
27
Mali
20
Burkina Faso
12
Gambia, The
19
Tanzania
US$
250
250
310
440
540
570
600
630
630
640
660
750
760
820
820
960
960
1020
1040
1040
5.5
LN Real GDP per capita (US$ of 2000)
LN Real GDP per capita, PPP (US$ of 2005)
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
5
1980
Logs
7
GNI per capita, Atlas method (Current $)
Ranking Country
1
Burundi
2
Congo, Dem. Rep.
3
Liberia
4
Ethiopia
5
Eritrea
6
Guinea-Bissau
7
Malawi
8
Sierra Leone
9
Niger
10
Rwanda
11
Gambia, The
12
Uganda
13
Madagascar
14
Mozambique
15
Nepal
16
Tajikistan
17
Central African Republic
18
Togo
19
Tanzania
20
Burkina Faso
13
Income Benchmark Tool Demo

Some definitions



Income choice
Atlas Method GDP for allocating countries
by income groups
Navigating the file:



Picking a country and benchmarks
Assumptions
Interpretation
14