THE ROLE OF MACROECONOMIC INSTABILITY ON ECONOMIC GROWTH RATE: THE CASE OF COLOMBIA, 1950-2009 Johanna Alejandra Pulido Pedraza [email protected]ABSTRACT The main objective of this paper is to investigate the role of macroeconomic instability on economic growth in Colombia, by utilizing a production function approach, over the 1950- 2009 period. Additionally, the role of openness and capital formation (both physical and human) on economic growth are investigated. In doing so, this paper has used modern time series techniques, such as unit roots, cointegration analysis and error correction models. Both the descriptive and econometric evidence show that the recurrent macroeconomic instability episodes seriously and negatively affected the growth potential of the Colombian economy during the 1950-2009 period. Empirical results also suggest that the growth in output is positively affected from physical and human capital formation but negatively affected from openness over the long term. Additionally, it is also found that in the long term, the output can adapt itself faster to the changes in macroeconomic instability and capital stock. Keywords: Macroeconomic instability, Economic growth, Human capital, Physical capital, Cointegration. JEL Code: E60, O40 1. Introduction Colombia has gone through a long way of restructuring not only socially, or politically, but also economically. Since all these are linked to each other, if one of them does not go on the right way, it will without a doubt affect the behavior of the other ones. Therefore, economic growth is only to be achieved when you find and create the right conditions that include many sectors and participants to work in unity, and in Colombian history, it is noticeable that these conditions were not so easy to balance or even to find. Throughout the history, Colombia had many different governments with many different development plans and objectives. These governments usually tried to enhance the growth
36
Embed
THE ROLE OF MACROECONOMIC INSTABILITY ON …teacongress.org/papers2012/PEDRAZA.pdf · THE ROLE OF MACROECONOMIC INSTABILITY ON ECONOMIC GROWTH RATE: THE CASE OF COLOMBIA, 1950-2009
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Sources: Cárdenas and Mercer-Blackman (2006); Bernal and Cárdenas (2003).
a. Includes parafiscal payroll taxes (contributions to ICBF, SENA, and CCFs),
contributions to social security (health, pensions, severance payments, and professional risks),
paid vacations, and mandatory bonuses.
A new reform was made by 2002, which tried to increase the formalization of
employment (It reduced overtime pay and made the workday and the workweek more
6 Nonwage costs are composed of the following: parafiscal contributions, which sum to 9 percent of
payroll, consisting of 2 percent for SENA (Servicio Nacional de Aprendizaje, the vocational training
program), 4 percent for ICBF (Instituto Colombiano de Bienestar Familiar, the family welfare
program), and 3 percent for the CCFs (cajas de compensación familiar, or family compensation funds), which are private funds involved in a host of activities including cash subsidies, recreation and
cultural activities, and unemployment insurance; contributions to severance funds and paid vacations;
health contributions, which rose from 7 percent to 12 percent in 1993 and to 12.5 percent in 2007; and
contributions to pensions, which rose from 6.5 percent to 13.5 percent in 1993 and to 15.5 percent in 2002, with an additional 1-percentage-point contribution for those with high salaries. Other sources
have estimated these percentages to be slightly different. These nonwage labor costs are calculated
using author-specific definitions of payroll taxes, mandatory bonuses, and so on. For example, according to the World Bank (2005), total nonwage costs were about 47 percent in the late 1980s and
rose to around 60 percent in 2004. ( Steiner, Clavijo and Salazar ,2000)
flexible, changed the apprenticeship contracts, which no longer involve parafiscal
contributions, and reduced the firing costs). Gaviria (2004) estimates that the reduction in
firing costs and overtime pay barely compensates for the increase in pension contributions
enacted at the same time.
3. A BRIEF REVIEW OF RELATED LITERATURE
In this part, first, there will be an explanation of the production function approach that
will be used in the empirical parts of the paper, afterwards there will be a review of the related
literature that uses the former approach. Subsequently, there will be a review of the literature
about the role of macroeconomic instability on economic growth, and how the political
economic factors affect the macroeconomic instability and growth. Finally, there will be a
review of some Colombian studies about the importance of macroeconomic instability and
related factors and their effects on the Colombian economy.
3.1. PRODUCTION FUNCTION APPROACH AND ITS APPLICATIONS: A
SELECTIVE REVIEW
The methodology of the production function is extensively used in empirical studies.
The Production function approach usually starts with the specification of a generalized Cobb-
Douglas.
One author who used this methodology in one of his works is Aschauer (1989). After
making a series of transformations, he proceeds to estimate the following equation for the
U.S. economy:
yt – kt = a0 + a1 t + a2 (nt – kt) + a3 (gt – kt) + a4 cut + ut (3.1)
yt – kt = a0 + a1 t + a2 (nt – kt)+ a31(k1t – kt) + a32 (k2t- kt) + a4 cut + ut (3.2)
where yt, kt, nt, gt, and cut are the logarithms of aggregate real production of the
private sector (Yt), the aggregate capital stock of non-residential (Kt), private sector
employment (Nt) of public capital stock (Gt) and the rate of capacity utilization (cut)7,
respectively. k2t and k1t represent the logarithm of the stock of public capital in two different
7 This term is included to take into account the influence of the economic cycle.
sectors of the economy8. a2, a3, a31, a32 and a4 are the output elasticities with respect to n, g, k1,
k2, and cu respectively. a0 + a1t is a term that seeks to collect technical progress. Finally, u t is
the error term in the econometric estimation.
The main aim of Aschauer (1989) is to analyze the role of different types of public
expenditures on output. After the estimation of the previous equation, Aschauer (1989)
reached to the proceeding conclusions: First, that the public investment of the government has
an important effect on economic growth9, and second, that the investment in infrastructure
(roads, water, electricity and gas) has a bigger impact on the productivity of the economy than
the government spending and the military public investments.
The main idea behind the work of Aschauer (1989) can be summarized as follows.
The level of productivity of the economy increases due a rise in the rate of return of private
capital10
that is created by the public investment, which in turn stimulates spending in private
investment and thus economic growth.
For another example of the application of the methodology of the production function,
we have Ghura‟s (1997) study, which investigated the factors that affected the economic
growth between 1963 and 1996 in Cameroon.
Ghura used the following Solow-Swan type of aggregate production function,
modified to include different types of capital stocks (private and government physical capital
stocks and human capital stock). The modified production function that Ghura uses in the
study is shown below:
Yt = Αt (Kp
t)α g
t)β (Zt)
Zt = Ht Lt (3.3)
Where Y is output; A represents technology; Kp
is private physical capital stock;
g is
the government physical capital stock; and Z is labor (L) adjusted for human capital
development (H).
8 Depending on the number of sectors it could include a k3t, k4t, etc.
9 That leads him to assert that a large proportion of the decline in the productivity of U.S. GDP during
the seventies can be explained by the decline in public investment spending. 10
The positive impact of public investment on private investment is greater than the negative
crowding-out effect, (See Aschauer 1989 for more detail).
The author finds that physical and human capital accumulation, as well as economic
policies, influenced the economic growth of Cameroon significantly. Additionally the private
and public investments have a positive influence on the economic growth. According to the
empirical results, an increase of 1 percentage point in private investment creates an increase
of 1.4 percentage point in the economic growth of Cameroon, this increment is larger than the
one experienced by Cameroon in the case that the public investment increases.
Ramirez (1998) studied the relation between public and private investment spending,
economic growth and productivity for the Mexican economy for the period from 1950 to
1990. Ramirez specified the public capital stock in a modified neoclassical production
function. The author finds that the growth in public and private investment spending has a
positive effect on the rate of productivity growth and the increment in government
consumption, in some cases, affects the rate of productivity growth negatively, which implies
that the composition of the government expenditure is important for the growth rate of the
economy and its productivity. The error correction models that the author performed gave a
positive feedback in terms of the simulation of the historical data on productivity growth for
the Mexican economy.
Khan and Kumar (1997) studied the role of public and private investment of growth for
95 developing countries for the period from 1970 to 1990 using a Cobb-Douglas production
function. The countries included in the paper account for 90% of GDP of developing
countries and this allows a broader consideration of the hypothesis that the public and private
investments have different effects on developing countries. The authors concluded that private
investment during the 1980s had a larger impact on growth than public investment, and that
the effects of public and private investment on growth and their rates of the return differ
significantly between the regions. For Latin America and Asia, the difference between these
effects is more apparent than for Africa, Europe and the Middle East. There is as well an
important difference across income groups, where the rates of return of public investment are
higher for low-income countries than for high-income countries. Finally, the authors conclude
that increasing competition and the exposure to foreign technology will have dynamic effects
on growth and that giving more importance to education will stimulate private investment and
a sustainable long-term economic growth.
3.2. INSTABILITY AND GROWTH
Ocampo (2005) provides an explanation about the macroeconomic stability concept,
by considering elements like price stability, fiscal policies as well as real economies that are
working well, debt ratios that are payable by the governments, public and private sector
balance sheets.
One of the most important indicators of lack of stability in an economy is the inflation
rate. A high inflation negatively affects the economy via capital accumulation and
productivity expenditures. A high inflation also raises the demand for dollars in the economy
at the cost of reducing the demand for domestic currency and the loss of the seigniorage
revenue (Ismihan, 2003).
So what happens when the level of instability of the economy increases? Well this
would affect the economic growth negatively through certain mechanism such as uncertainty
that decreases the rate of productivity of the economy and also the rates of private investment
and will create a wave of capital flight that will decrease the capital accumulation (Ismihan,
2003).
So as noted before an economy needs stability to be able to enhance a long-term
growth pattern, and to induce the investment (public and private as well).
3.2.1. Empirical Evidence
Cardoso (1993) analyzed the private investment in Latin America and its reaction to
instability, growth and to real exchange rate depreciation as well as to the terms of trade. The
author uses quadrennial panel data for Argentina, Brazil, Chile, Colombia, Mexico and
Venezuela for the period from 1970 to 1985. Cardoso finds that the variation of private
investment share in output is explained as 74% by the growth, the share of public investment
in GDP, and the log of terms of trade.
Also the author confirms the hypothesis of complementarity between the private and
public investments, and that the improvement on the variable of terms of trade affected the
investment positively. Additionally, Cardoso found that the real exchange rate and the real
rate of depreciation had no effect on investment. Finally, and most importantly, Cardoso
(1993) found a negative effect of economic instability on private investment.
A seminal paper by Fisher (1991) investigated the connection between the
macroeconomic policies and growth. The author came to the conclusion that the stability in
the macroeconomic environment is necessary for growth and for it to be sustained for longer
periods. Moreover, the economic strategies that the government pursue and the size and role
of the government (in the provision of physical and social infrastructure especially for human
capital) is crucial.
Ismihan (2003) conducted an empirical study for the case of Turkey for the period
from 1963 to 1999 by including variables like real GNP, real private fixed investment, real
public investment, real public fixed infrastructural investments and the macroeconomic
instability index. He reached to the conclusion that the two types of public investment have
asymmetric effects on the macroeconomic performance and that the instability that affected
the macroeconomic environment, had a negative effect on the capital formation and on the
economic growth. Ismihan (2003) also concluded that the macroeconomic instability is a
severe problem which damages the relations between public and private investments in the
long-term.
Ismihan (2009) provides another empirical study about the role of the macroeconomic
instability on the potential growth rate of the Turkish economy from 1960 to 2006. This study
uses the production function approach and concludes that the Turkish economy suffered from
a significant output loss during the chronic instability episodes, between 1970s and 2001.
However, Turkey from 2002 to 2006 had a high growth mainly due to the reduction in
macroeconomic instability. Finally, he concluded that if the Turkish government wants to
continue high growth rates it is necessary to speed up both human and physical capital
formation while maintaining stability.
Another example of the relationship between macroeconomic stability and economic
growth is given by Sanchez-Robles (2006) who studied relations such as the macroeconomic
instability (composed by inflation, public deficit and various types of public expenditure as a
share of the GDP) and market distortions for the Spanish economy for the period from 1962
to 1995. The findings of the author can be summarized as follows:
Investment/GDP is highly correlated with the economic growth in the short and long
run (this indicates that the capital accumulation has an important role in the economic
growth of Spain in the last decades),
The inflation rate affects the economic growth negatively,
The total public expenditure as share of the GDP affects the economic growth
negatively in the long-run (this is especially the case of unproductive expenditure),
The public deficit/GDP also affects the economic growth negatively,
Distortions in some markets (especially foreign exchange and oil market) appear to
affect the economic growth negatively,
As a final conclusion the macroeconomic instability and the markets distortions have a
negative correlation with economic growth. Sanchez-Robles give the advice that the
macroeconomic stability and the liberalization of markets are necessary for the
Spanish economic growth.
3.3. THE ROLE OF POLITICAL ECONOMY FACTORS IN
MACROECONOMIC INSTABILITY AND GROWTH
Instability in political and economical environment has important effects over the
economic growth of a country. A paper by Rodrik (1999) has pointed out that democracy is an
important variable of economic growth. When a country suffers from lack of democracy, it
affects the output of the country negatively. However, when the citizens force the government
to be more democratic, this will encourage better policies that could benefit the society.
Rodrik (2000) analyzes wether it is wise to use the integration policies as a development
strategy for the developing countries. Even integration has positive sides for economic growth
in developing countries, it also brings within highly demanding institutional prerequisites (the
institutional changes are expensive and these changes involve the expenditure of scarce
human resources, administrative capabilities and political capital that could actually take these
capitals away from more urgent priorities). Rodrik concludes that “openness is not an
adequate substitute for a developing strategy” (2000: P2) and that even many economists
know that trade comes with many gains, these gains actually tend to be small. Even with the
evidence on trade protection should not be preferred for trade liberalization. The author
suggests that a strategic use of international trade and capital flows is necessary (especially
the developing economies should open their capital accounts in an organized and progressive
manner) as part of a strategy of development, not as substitute for it.
Institutions and their role in economic growth has become an important subject of study
by many economists and as Douglass North defines “efficient institutions motivate self-
interested individuals to act in ways which contribute to collective welfare” (qtd. Szirmai,
2012, pp. 35).
For instance, Baily (1994) suggests that institutions play an important role in the growth
and in the stagnation of growth in the economies. Baily explains that in Latin America the
capital accumulation did not guarantee economic growth because some of the investments in
infrastructure never became operational (steel facilities specially) and in human capital
accumulation even there were bigger investments, the skills of the work force sometimes are
not well utilized because of inefficient companies that in some cases are regulated by the
government. In the study, Baily analyzed five developing economies (Argentina, Brazil,
Colombia, Mexico and Venezuela) in comparison with developed economies and their
performance in four industries (steel, processed food, retail banking and telecommunications).
In this study, the author concluded that a shortage of capital was not the only reason for low
productivity in the afore-mentioned countries: in some cases the investments were wasted or
given to a particular sectors instead of diversifying them. Baily also found that the shortage of
skilled production labor was not a significant reason for the low productivity (what is
necessary is that the companies should have trained their employees better). Finally, Baily
affirms that if the Latin American countries pay more attention to creating better and more
efficient institutions this could create a better long-term growth.
An important cause of instability is the political institutions. If a country has many
political parties; this will have a negative effect over the macroeconomic stability (because of
the large transfers of funds from the government to the parties as well as the diversity of ideas
and plans launched).
Reinhart (2004) found as well, that the more democratic the institutions are, the better
the perception of the risk of lending to that country by institutional investors is.
According to Rajan and Zingales (2003), trade openness is a mechanism for limiting
the redistribution of the wealth towards the elites because if a country opens its trade, it will
raise the cost of the monetary policies (inflation) and it will be harder for particular
individuals to justify the self distribution.
3.4. ECONOMETRIC STUDIES FOR THE COLOMBIAN ECONOMY: AN
OVERVIEW
Macroeconomic instability, poor public investment, inequality in the distribution of
wealth and corruption within the political institutions (where the primary budgets are
managed) are frequently the problems related to Latin American countries.
One author who studied the role of corruption in Colombia was Rojas (2004), who
concluded that the corruption is one of the problems that must be fixed in a developing
economy like Colombia, because it has a negative effect over the economic growth. A recent
example is that the last mayor of Bogota (Colombia‟s Capital) was involved in the so-
publicized hiring carousel (carrusel de las contrataciones) that cost to the city more than 2
billions of pesos (RCTV) and the destruction of the roads and also the dismissal of Samuel
Moreno from the administration (and the new relocation of contracts that were already paid to
not so ideal companies for the development of infrastructure in Bogota) which ended up
reducing the output of Bogotá that is one of the major contributors of the economic growth in
Colombia .
In addition, there are studies about the role of the infrastructure on economic growth
and one example is the paper of Strauch (2002) for the Republic Bank. Strauch concludes that
the total public investment has no significant impact on economic growth. However, if there
is an increase in public investment in the areas of electricity, gas, water, education, mining
and manufacturing, it is possible to obtain higher levels of production.
Unlike other studies, public investments in infrastructure, especially the one directed to
construction of roads, did not have a strong positive impact on economic growth; on the
contrary, it was slightly negative. The above conclusion may be that in this study, unlike other
studies, public investments in roads correspond to monetary disbursements made by the
government and not to the number of existing roads.
A study that researches the effects of external public debt is the one made by
Salamanca and Monroy (2008) that used a TAR non-linear model for the period from 1994 to
2007 to explain how the external debt of Colombia could affect the private investments. The
authors came to the conclusion that there is no linearity between them and that there is an
inverse relationship between the growth of investment and the growth of the external public
debt. They also provide evidence that the long-term inverse relationship between the variables
is symmetric and that in the short-term the relationship is asymmetric. Finally, they concluded
that recessions affect the investments in a more permanent way, when the economy is in a
state of external excessive indebtedness.
In the topic of macroeconomic instability and its effect on growth in the case of
Colombia, we can find the research made by Cardenas and Urrutia (1995). This study
concludes that, for the Colombian economy, the macroeconomic stability is a necessary,
though not a sufficient, condition to achieve social progress. The best instruments for reaching
economic stability are the institutions and policies and that the social benefits generated by
fiscal and economic stability are worth the costs for being able to reach it.
4. THE ROLE OF MACROECONOMIC INSTABILITY ON GROWTH RATE IN
COLOMBIA: AN EMPIRICAL INVESTIGATION
In this section, the empirical model is estimated for the Colombian economy for the
period of 1950 to 2009. Initially unit root tests are conducted to determine the order of the
integration of the variables. Both long-run and short-run analyses are performed by using
cointegration and error correction models, respectively.
4.1. THE MODEL
To investigate the role of macroeconomic instability on economic growth rate in
Colombia between 1950 and 2009, its used the approach of the production function and
specified the following Cobb Douglas model11
:
Yt = e Θ
0 + Θ
1 Z
1t + Θ
2 Z
2t +…..+ Θ
m Z
mt Ktα Ht
β (4.1)
where Yt represents real GDP, Kt represents capital stock, Ht represents human capital,
the parameters α (β) represent the output elasticity of physical (human) capital and Zt (i=1,
11
For more detail, see Ismihan (2009).
2,…., m) may represent openness, research and development expenditures12
, and
macroeconomic instability which may affect the level of total factor productivity.
Following Ismihan (2009), its obtained the below log-linear model: