journal review_growth acceleration

8/2/2019 Journal Review_Growth Acceleration

1/9

A review of Journal Article :

Growth Accelerations

Ricardo Hausmann, Lant Pritchett, Dani RodrikJournal of Economic Growth, Vol. 10, No. 4 (Dec., 2005), pp. 303-329

Reviewed by: La Ode Sabaruddin

Introduction

This journal review is based on a Development Economics class assignment in

Graduate Program in Economics University of Indonesia, even semester, academic year

2010/2011 instructed by Padang Wicaksono, Ph.D. It is to review a journal article

relevant to a topic discussed in the course. This assignment reviews article Growth

Accelerations by Ricardo Hausmann, Lant Pritchett, Dani Rodrik (2005) on the

Journal of Economic Growth. In this review I respectfully refer to Hausmann, Pritchett

and Rodrik (2005) as HPR.

Issues relating to the growth acceleration have been the subject matter of

discussion in economic policy. It would be interesting to examine the sources of such an

impressive acceleration in some countriess growth rate due to why some countries

grow faster than others. In developing countries where rapid economic growth has

commonly become a national goal, analysis of the sources of growth assumes special

significance not only because it helps to find out what has and what has not been

important in the growth which has already occurred, but also because of the obvious

implications it has for the macroeconomic strategy and policies that affect the future

growth - its rate as well as pattern. An analysis of the factors leading to the significant

acceleration in the economic growth would be useful in assessing the possibilities of

a further acceleration in the future. HPR article may become a preferable article in order

to explain the accelerating process of economic growth.

Motivation for the Paper

I strongly motivated on reviewing of HPR article (2005) based on the fact that

this paper is a seminal paper which identify significant determinants of growth

accelerations. HPR use an unconventional approach to identify drivers of differential


2/9

growth. Instead of running cross-sectional or panel estimations as in Barro (1991) or

Islam (1995), HPR begin by identifying turning points in growth episodes and then

examine their determinants around these turning points. This approach differs to prior

studies which assume that a single linear model is appropriate for all countries at all

times. HPR approach has attracted much attention and some recent papers have

followed their approach. As of September 2010, the article had accumulated more than

50 citations in the Web of Science (Guo Xu, 2011). HPR study has mainly considered

as a pioneer study in this field regarding their unconventional approach. They examine

whether political regime changes, external shocks, and economic reforms precede

growth accelerations.

Argument of the Paper

Numerous study have tried to assess the pattern of accelerating process on

economic growth relates to impact of political, legal, and economic institutions which

used variety of econometrics approaches (Barro, 1991; Barro and Sala-i-Martin, 1992;

Hall and Jones, 1999; Acemoglu et al., 2001; Islam, 1995; Caselli et al., 1996; Temple,

1999, Durlauf 2003, and Easterly 2003). The usefulness of the growth regression

framework is questionable, however, as it assumes that a single linear model is

appropriate for all countries at all times. Easterly et al. (1993) and many others

confirmed later pointed out that growth performance tends to be highly unstable. Very

few countries have experienced consistently constant growth rates over time. Pritchett

(2000) documents, for instance, that the variation in growth rates within countries is

large relative to both the average growth rates and the variance across countries.

Likewise, Jones and Olken (2005) report that no less than 48 countries have experienced

one or more structural breaks in their economic development. These breaks lead to very

distinct growth patterns. Whereas some countries have sustained long periods of

growth, others have experienced rapid growth followed by stagnation or even crisis.

Still others have faced continuous stagnation or steady decline.

One promising research strategy is to examine the economic, political,

institutional, and policy conditions that accompany changes in growth patterns. HPR


3/9

contributes in this field through identifying the turning points in growth experience and

examine whether political regime changes, external shocks and economic reforms

precede growth accelerations. HPR first employ a filter rule to identify sudden periods

of growth accelerations. By then examining changes in policies and plausible variables

around these turning points.

The Main Claims of the Paper

HPR paper demonstrates four important points :

First, growth accelerations are fairly frequent occurrence. HPR identify 83 periods of

110 countries accelerated growth that are sustained for at least 8 years for the period of

19571992.

Second, growth accelerations tend to be correlated with increases in investment and

trade, and with real exchange rate depreciations.

Third, political-regime changes are statistically significant predictors of growth

accelerations, while the nature of other determinants depends to some extent on whether

the acceleration is sustained into the longer term or not. External shocks tend to produce

growth accelerations that eventually fizzle out, while economic reform is a statistically

significant predictor of growth accelerations that are sustained.

Fourth, growth accelerations tend to be highly unpredictable: the vast majority of

growth accelerations are unrelated to standard determinants such as political change and

economic reform, and most instances of economic reform do not produce growth

accelerations.

Assesment of extent which the Claims are Justified

In their paper, HPR identify the periods of accelerated growth using the

following filter. For each country (with more than 1 million inhabitants and more than

20 available observations) the logarithm of real GDP per capita (taken from Penn World

Table 6.1) is regressed on time for every eight-year period (n=7). That is,


4/9

Wherey denotes real GDP capita and tis time. The estimated parameter, gt,t+n,

is taken as a proxy for the average growth rate over the period tto t+n and labeled the

least squares growth rate. To qualify as a growth acceleration, the least squares

growth rate must be at least 3.5% per annum. Furthermore, it must be at least 2

percentage points higher than in the previous eight years. Finally, to rule out episodes of

full economic recovery, the level of real GDP must be higher at the end of the

acceleration than in all years before the acceleration. In cases in which consecutive

years qualify to be the start of a growth acceleration, the year is chosen with the highest

F-statistic of a piecewise linear (or spline) regression with the break at the relevant year.

HPR allow for the possibility that an acceleration is followed by another acceleration as

long as the second acceleration starts at least five years after the first one.

By definition, a growth acceleration has occurred if and only if :

gt, t+7 3.5ppa Growth is rapid(1)

gt, t+7 2ppa Growth accelerates (2)yt+7 max(yi), i t Post-growth output exceeds pre-episode break(3)

This filter yields 83 growth accelerations for 110 countries (as shown in Table 1: HPR

2005, 307-308). The unconditional probability that a country will experience a growth

acceleration sometime during a decade about 25 percent (as shown in table 2: HPR

2005, 310). Of the 110 countries included in the sample, 60 had at least one acceleration

in the 35-year period between 1957 and 1992-a ratio of 55 percent.

As a precursor of their regression analysis, HPR investigate variables to be

associated with the start of a growth transition using six variables (terms of trade

changes, imports, exports, inflation, investment, and real exchange rate). The results as

follows (shown in Table 6, HPR 2005, 316) :

First, Correlates of growth around the date of growth transition (t-1, t, and t +1 where t

is the date of the acceleration compared to the value of that same variable during seven

previous years) show that growth acceleration coincide with an increase in imports,


5/9

exports, investment, and real exchange rate. Meanwhile, terms of trade changes and

inflation have positive correlation but not statistically significant at the time of the

transition.

Second, Correlates of growth during the 8-year growth acceleration produce similar

results except for real exchange rate. Real exchange rate changes are no longer

statistically significant.

By these results, HPR denotes that growth acceleration require more investment, more

exports, and a more competitive real exchange rate.

To investigate the predictors of growth acceleration, HPR using three types of

predictors :

(i) External shocks (tot_thresh90). Growth accelerations may be triggered byfavorable external conditions, and HPR therefore include a terms-of-trade

dummy, which takes the value 1 whenever the change in the terms of trade

from year t-4 to tis in the upper 10 percent of the entire sample.

(ii) Economic reform. To quantify a change in economic policy, HPR relyprimarily on an index provided by Wacziarg and Welch (2003), which

incorporates a number of structural features (e.g., presence of marketing boards

and socialist economic regimes) and the macroeconomic environment (e.g.,

presence of a large black-market premium for foreign currency), in addition to

tariff and non-tariff barriers to trade. The variable used (econ_lib) is a dummy

that takes the value of 1 during the first five years of a transition towards

openness. In addition, HPR also includes financial liberation,finance, which

is a dummy for the first 5 years of a financial liberalization episode.

(iii) Political regime changes (reg_change) are proxied by a dummy that takes avalue of 1 in the 5-year period beginning with a regime change as recorded in

the Polity IV dataset (Marshall and Jaggers 2002), where a regime change is

defined as a three-unit change in the Polity score (or as a regime interruption).

In addition, HPR includes some additional political variables, poschange (a

movement towards greater democracy), negchange (move towards greater

authoritarianism), lead_death (a dummy for the 5 year priod starting with a

political leaders death), tenure (interanction term between lead_death and the


6/9

length of the tenure of the dying ruler), war_end (a dummy for the 5 year

period beginning with the cessation of an armed conflict), and civil_war (a

dummy for the 5 year period beginning with the ending of an armed civil war).

HPR starts analyze the predictors of growth acceleration with a simple bivariate

examination of the relationship between growth accelerations and its potential

determinants. To give the determinants time to make their effect felt, HPR applies up to

a maximum of 5 yearss lag between a change in determinant and the growth

acceleration. The predictability of the growth acceleration demonstrates some points

(shown in table 7, HPR 2005, 320) :

First, predictive power of the economic reform variable (econ_lib) shows that only 14.5

percent acceleration are associated with economic liberation. In other words, 85.5

percent of growth accelerations are not preceded or accompanied by liberalizations.

Second, around half of growth accelerations are preceded by political-regime changes,

but only a tiny proportion of political regime changes are followed by growth

accelerations.

Third, the relationship between growth accelerations and positive terms of trade shock

is quite weak (only 5 percent of positive terms of trade shock are followed by growth

accelerations).

By then, HPR continues their analysis through estimates specification using a

probit model. HPR main baseline specification is :

1)=(0+ 1tot_thresh90it+ 2econlibit+ 3regchangeit+ T)

where episodeit is 1 if there is a growth acceration within (t-1, t+1) in country i and 0

otherwise. tot_thresh90it, 2econlibit, and3regchangeit are 1 in (t, t+4) following an

event at t. T are time dummies to capture shocks common to all countries and is the

cumulative distribution function of the standar normal distribution.

All specifications are estimated using a probit model, and the results are as follows

(shown in table 8, HPR 2005, 322) :


7/9

First, terms of trade shock has statistically significant coefficient. A terms of trade

shock changes increases the probability of experiencing a growth acceleration by 4.4

percentage points.

Second, political regime change has also statistically significant coefficient. A political

regime changes increases the probability by 5.3 percentage points.

Third, economic reform does not have a statistically significant impact on growth

acceleration, although its estimated coefficient is (as expected) positive.

By then, HPR classified acceleration into un-sustained accelerations and

sustained accelerations beyond eight year horizon. Acceleration is considered as

sustained episodes if the growth rate remained above 2 percent in years (t+7, t+17), and

those where the growth rate fell below the 2 percent threshold un-sustained episodes.

The results (shown in table 12, HPR 2005, 327) as follows :

First, terms of trade shock are conducive only to un-sustained episodes; it has no

predictive power over sustained episodes.

Second, Economic reform has a statistically and quantitatively significant impact on the

likelihood of sustained accelerations.

Third, Financial liberalizationss positive impact in confined to unstained accelerations

Fourth, Positive political change (democratization) has a significant impact on sustained

episodes but not unsustained episodes.

Finally, HPR concludes that these results strongly suggest that unsustained and

sustained growth tend to be triggered by different condition. Thus, growth acceleration

tend to be highly unpredictable due to the vast majority of growth acceleration are

unrelated to standar determinants. Those determinants do a very poor job of predicting

the turning points.

Weaknesses of the paper

Most critical aspects in HPR article are methodological issues of turning points

studies. HPR first employ a filter rule to identify sudden periods of growth

accelerations. By then examining changes in policies and plausible variable around

these turning points, HPR seek to isolate robust relationship between changes in policy


8/9

and growth trajectory. While the longitudinal approach of HPR appears particularly

appealing for testing theories beyond averages, it may faces some methodological

weaknesses, such as omitted variables, endogeneity, and measurement errors. Ideally,

this concerns could be addressed by randomized controlled trial (Banerjee and Duflo,

2008). To disentangle the effect of policies from shocks, one would randomly assign

countries to treatment and control groups, and then manipulate only a certain policy

variable in the treatment group. It is hoped that, given the exogenous ex-ante group

assignment, shocks and other unobserved confounds would be balanced across both

groups. Any differential in growth performance across groups would then be causally

attributed to the treatment.

Turning-point studies are vulnerable to problem arising from the poverty of data.

Unlike crosssectional studies, turning point studies require long time-series which are

often unavailable. If most of the missing values are either dropped or coded zero as is

done in HPR article, selection biases could occur, as missing values are often correlated

with country characteristic. Turning point studies focusing on rare events are

particularly prone to missing values, as the approach often involves the loss of valuable

observations. Some critical comments due to these methodological issues were came

from Jong-A-Pin and De Haan (2008), and Xu (2011).

References

Banerjee, A.V., and E. Duflo. 2008. The Experimental Approach to Development

Economics. NBER Working Paper 14467, National Bureau of Economic

Research, Cambridge, MA.

Barro, R.J. 1991. Economic Growth in a Cross-section of Countries. The Quarterly

Journal of Economics 106(2): 407-443.

Hausmann, Ricardo, Lant Pritchett, and Dani Rodrik. 2005. Growth Accelerations.

Journal of Economic Growth 10(4): 303-329.

Islam, Nazrul. 1995. Growth Empirics: A Panel Data Approach. The Quarterly Journal

of Economics 110(4): 1127-1170.

Jong-A-Pin, R. and J. de Haan. 2008. Growth Accelerations and Regime Changes: A

Correction.Econ Journal Watch 5(1): 5158.

Pritchett, Lant. 2000. Understanding Patterns of Economic Growth: Searching for

Hills among Plateaus, Mountains, and Plains. World Bank Economic Review

14(2): 221-150.


9/9

Wacziarg, R. and K. H. Welch. 2003. Trade Liberalization and Growth: New

Evidence. NBER Working Paper10152, National Bureau of Economic Research,

Cambridge, MA

Xu, G.. 2011. Growth Accelerations Revisited.Econ Journal Watch 8(1): 3956.

journal review_growth acceleration

Documents