Transcript Alejandro Werner

Macroeconomic Volatility and Growth:
Evidence and Policies
Alejandro Werner
March 2006
Macroeconomic volatility in Latin America has been relatively
high. Evidence suggests that this has been a negative factor for
growth in the region.
Per Capita GDP Growth Volatility
(Std dev., 1970-2003, %)
4.5
Per Capita GDP Volatility and Growth
(Avg. and std dev., 1970-2003, %)
Average growth
1970-2003
4.1
4
4.0
3
3.5
3.1
2
3.0
1
2.5
0
1
2.0
-1
Latin America
Rest of the World
5
9
13
17
Standard Deviation
1970-2003
Source: Alan Heston, Robert Summers and Betina Aten, Penn World Tables Version 6.1. For 2001-2003, per capita GDP growth data is
from World Economic Outlook, IMF, September 2005.
Due to slow growth in Latin America, per capita GDP deteriorated in
comparison to other economies. A clear exception to this regional
trend since the 2nd half of the 80s is Chile.
Asia: Per Capita GDP Ranking
(Position in a 98-country sample)
(Position in a 98-country sample)
Taiwan
Thailand
2002
2002
1996
1990
1984
1978
1972
1966
46
Singapore
1996
41
Korea
1990
36
Phillipines
1984
31
Japan
1978
26
H. Kong
1972
21
1960
Position (1=highest
16
China
1966
Argentina
Brazil
Chile
Mexico
Venezuela
1
6
11
16
21
26
31
36
41
46
51
56
61
66
71
1960
11
Position (1=highest) .
Latin America: Per Capita GDP Ranking
Source: Alan Heston, Robert Summers and Betina Aten, Penn World Tables Version 6.1. For 2001-2003, the GDP per capita
growth data from World Economic Outlook, IMF, September 2005 is used.
Chile carried out a broad reform agenda that delivered sustained
growth and reduced macroeconomic volatility. However, Chilean
economic takeoff happened long after the first wave of reforms.
Chile: Reforms, Per Capita GDP and Growth
Volatility
(Thousand dollars of 1996, and std. dev. in %)
1st
2nd
Reforms Reforms
GDP
10.5
Average Annual Growth and Volatility
of Chilean Per Capita GDP
(Percent)
12
Volatility
9.5
Period
10
8.5
GDP
7.5
6
6.5
Volatility
8
1960-1973
1974-1980
1981-1984
1985-1990
1991-2005
Average
Growth
1.9
1.5
-2.6
3.9
4.0
Growth
Volatility
4.3
8.6
7.9
2.4
3.0
4
5.5
2
4.5
2000
1996
1992
1988
1984
1980
1976
1972
1968
1964
0 Source: World Penn Tables, IFS and Larraín and Vergara, “Chile en
1960
3.5
Pos del Desarrollo: Veinticinco Años de Transformaciones Económicas”
in La Transformación Económica de Chile, Larraín and Vergara (ed.)
Macroeconomic volatility in Latin America is explained by
external and domestic sources. Endogenous sources have
magnified the effect and scope of exogenous factors.
a) External
• Terms of trade shocks
• Adjustments in global financial markets
Exogenous
b) Domestic
• Political instability
• Fiscal and monetary policies
• Financial system
Endogenous
Latin America has been subject to larger external disturbances than
other regions. Depending on exports’ concentration, terms of trade
shocks have been a leading source of instability.
Terms of Trade and Capital Flows Volatility
(Std. deviation)
Terms of trade (growth rate)
16
Int`l capital flows (% of GDP)
Terms of Trade Volatility
(Std. deviation)
4.5
4.0
3.5
12
3.0
2.5
8
2.0
1.5
4
1.0
0.5
Source: Inter-American Development Bank.
Other E.
Asia and
Pacific
South
Asia
East Asia
Industrial
Countries
0.0
Latin
America
0
Venezuela
Ecuador
Brazil
Bolivia
Colombia
Peru
Nicaragua
Guatemala
El Salvador
Chile
Uruguay
Mexico
Costa Rica
Paraguay
Honduras
Argentina
Panama
50.63
35.10
21.21
19.31
16.70
14.74
14.29
13.79
13.79
12.81
12.03
11.96
11.52
11.43
10.17
10.05
7.85
0
20
40
60
Souce: IDB Economic and Social Database (GDP) and World
Bank World Tables (terms of trade).
Also, the region has shown a considerable political
instability.
•
Latin America ranks second to last (only above Africa) in terms of
rule of law and corruption control.1/
•
Out of 133 presidential administrations in Latin America between
1945 and 1982, 51 ended in coups.2/
•
Electoral volatility in sixty percent of Latin American countries
exceeded that of the most volatile European country by an average
of 126%.1/
•
Latin America ranks as the region with the greatest political
gridlocks between 1985 and 1994, more than double any other
score.1/
•
Less than 40% of Latin Americans are very or partly satisfied with
democracy in their countries.1/
1/ Inter-American Development Bank (IDB) (2000): Development: Beyond Economics. Washington, D.C.: IDB.
2/ Ames, B. (1987): Political Survival: Politicians and Public Policy in Latin America. Berkeley: University of California Press.
Fiscal policy in Latin America has been strongly procyclical,
thus amplifying macroeconomic volatility from exogenous
sources.
Index of Cyclicality of Fiscal Policy
0.50
0.40
0.30
0.20
0.10
0.00
-0.10
-0.20
-0.30
-0.40
Uruguay
Peru
Iran
Paraguay
Venezuela,
Guatemala
Argentina
Philippines
India
Chile
Brazil
Tunisia
Costa Rica
Mexico
Bolivia
Egypt
Malaysia
Indonesia
South Africa
Japan
Greece
Turkey
Colombia
China
Italy
Korea
Australia
Canada
Jamaica
U.S.
Spain
France
U.K.
Denmark
Finland
-0.50
Souce: Kaminsky, Reinhart and Végh (2004)
The ICFP is a composite index of correlations of the cyclical components of fiscal policy instruments (Real Central Government Expenditure and the Inflation
Tax) and real GDP (HP Filter) and a measure of the amplitude of Real Central Government Expenditure Cycle captured by the difference in growth of real
central government expenditure in "Good" and "Bad" Times. The amplitude of the real central government expenditure cycle is transformed to a (-1,1) indicator.
Following the liberalization of the 80s, the financial system
emerged as a mechanism that also amplified these exogenous
volatility sources.
 Two characteristics of the financial systems of the region have
accounted for this role:
• Low global integration
• Small size and inefficiency
 Underdeveloped financial systems have fueled crises in LA
• Their lack of global integration has prevented them from
providing liquidity at difficult times
• Small size and inefficiency have prevented them from allocating
funds to distressed but profitable firms
 Financial systems have helped turn balance of payments crises into
macroeconomic crises across the region
In fact, policy volatility (endogenous sources) accounts for a
large fraction of overall macroeconomic volatility and its costs
across countries.
Estimated Impact of Volatility on Growth*
(Percent)
0.10
-0.30
-0.70
Policy Volatility
Terms of Trade Volatility
-1.10
Latin
America
East
Asia
South
Asia
Other E.
SubAsia and Saharan
Pacific
Africa
Estimated Impact of Volatility on Growth*
(Percent)
Bolivia
Venezuela
Ecuador
Nicaragua
Argentina
Peru
Chile
Salvador
Uruguay
Brazil
Mexico
Colombia
Costa Rica
Guatemala
Paraguay
Honduras
Panama
Terms of Trade
Volatility
Policy Volatility
-3.8 -3.3 -2.8 -2.3 -1.8 -1.3 -0.8 -0.3 0.2
Source: Inter-American Development Bank.
Source: Inter-American Development Bank.
*Note: For each region, numbers represent the impact on growth of volatility in excess of that observed in industrial countries.
The sources and amplifiers of macroeconomic volatility in Latin
America reflect a poor institutional development.
 Therefore, efforts should focus on developing and strengthening
those institutions aimed at delivering sound and time-consistent
policies:
 Reduce the effects of external volatility
 Curbing those channels that amplify exogenous shocks
 Macroeconomic policy
 Financial sector
Specifically, the following steps can be effective in reducing
overall macroeconomic volatility
 Reduce the impact of external shocks
• Diversify exports
• Choose an adequate exchange rate regime
• Develop stabilization funds and other forms of self-insurance
• Pursue trade and investment treaties
• Creates long-term links to developed foreign institutions
• Increases the resilience of external sector (Calvo, 2005)
• Exports risks by diversifying ownership
 Macroeconomic policy
• Well-defined monetary policy objectives (transparency)
• Prudent fiscal targets and rules for unexpected shocks
• Reduce public debt sensitivity to external shocks
 Financial development
• Promote financial global integration
• Deepen financial markets by promoting domestic saving
It is worth noticing that “growth acceleration” episodes have
been preceded by changes in the institutional framework.
Growth Acceleration.- An increase of per capita GDP growth of 2pp or
more, sustained for at least 8 years and the post acceleration growth
rate has to be at least 3.5% per year (Hausmann, Pritchett, Rodrik, 2004).
While growth accelerations tend to be highly unpredictable, some
common factors can be found:
• Sustained accelerations are statistically predicted by political
changes towards democracy (if only developed countries are
considered, shifts towards autocracy have a positive effect).
• Positive external shocks produce only temporary accelerations.
• Sustained growth accelerations are statistically predicted by
economic reforms.
An interesting question for future research is whether such acceleration
episodes have been characterized by volatility drops.
Concluding Remarks
 Even successful cases as Chile’s show that it takes time (and
patience) for a growth-accelerating institutional change to yield
results. So, we should keep working on the long-term institutional
development.
 In the meantime, second-best policies designed to operate under a
weak institutional environment should be strongly considered.
 For example, linking our institutions to more developed external
counterparts through treaties or setting clear laws for policy
actions can deliver positive results in the medium–term.
 This would also shield the long-term institutional development
process from volatility bouts that could jeopardize it.
Macroeconomic Volatility and Growth:
Evidence and Policies
Alejandro Werner
March 2006