Does growth matter for government and external debt in the young democracies of South America? Yes it does!

Politics and Economy


South America has been known for political transitions from dictatorships to more democratic regimes, macroeconomic instability (some countries experienced debt crises and severe episodes of high in‡flation), delayed stabilisation processes (exempli…ed by the ‘"lost decade’") and, on a more positive note, no come back to less democratic regimes as in the more distant past. Moreover, the region has been known for a certain, relatively above the average, degree of economic inequality.

Against this rather eventful background, and also with the current debt crisis affecting some (Mediterranean) European countries in mind, we argue that economic growth has played an important role in reducing government and external debt in the young South American democracies during the 1970-2007 period. Essentially, we suggest that consistent economic activity is paramount in keeping debt under control, or even in reducing it to lower levels (i.e., during booms, or periods of consistent economic activity, governments not only collect more taxes, but also spend less on welfare, which can lower the debt).

Loosely speaking, in times of a renewed wave of populism taking place in some parts of South America, an environment which incentivises economic activity is perhaps a better option than the current contempt for property rights taking place in countries like Argentina and Bolivia, a contempt that tends to generate economic uncertainty and instability, lower investment, and consequently reduced economic activity with its associated higher government debt. Needless to say that lower growth rates are troublesome for a number of other reasons, e.g., Dollar and Kraay (2002) suggest that "growth is good for the poor" because it helps to reduce income inequality.

A look at the data

The data we use covers the period between 1970 and 2007, and all nine South American countries which transitioned from political dictatorship to full democracy at some point in the late 1970s (Ecuador), 1980s (Argentina, Bolivia, Brazil, Chile, Peru and Uruguay), and early 1990s (Guyana and Paraguay).

The variables for government and external debt are the share of general public debt to GDP (govdebt), from the Historical Public Debt Database provided by the IMF, and the share of external debt to GDP (extdebt), from the World Development Indicators, which is provided by the World Bank. With information on these two variables for debt we create an extra index for debt, government1, which roughly corresponds to the mean of those above-mentioned series. Data on economic growth (growth) come from the Penn World Table.

To illustrate the behaviour of these variables for government and external debt over time in the region, in Figure One we plot, in clockwise fashion, all these averaged series against time. This eyeball evidence shows that these country averages increased moderately during the late 1970s and rather dramatically in the early 1980s, which roughly coincides with the implementation of more democratic regimes in the region (alternatively it can also coincide with the end of those political dictatorships). Moreover, the dramatic increase in government and external debt in the early 1980s coincides with the hyperin‡ationary episodes that some of those countries experienced at the time (e.g., in Argentina, Bolivia, Brazil, Peru and Uruguay).

On the other hand, the debt series present a reasonably consistent reduction from the 1990s onwards, which suggests that some time after redemocratisation, and with the macroeconomic stabilisation and institutional reforms taking place in most of those countries (e.g., …fiscal rules, a certain degree of central bank independence and the end of the import substitution model), the size of debt has actually decreased in the continent.

In addition, to illustrate the general macroeconomic conditions in the continent, we plot the economic growth averages over time, and we can see that growth rates displayed even negative fi…gures in the 1980s (the "‘lost decade’"), which coincide with the sharp increase in public and external debt seen in the fi…rst three panels. However, those averages have been displaying a more encouraging positive trend from the 1990s onwards, which broadly coincides with the macroeconomic stabilisation taking place in the 1990s, and also speci…cally with the reduction in debt that the region has experienced recently.

Figure 1: Government debt, external debt, government and economic growth, South America, 1970-2007. Sources: IMF, World Bank and PWT …les.
Government debt, external debt, government and economic growth, South America


Moreover, in an attempt to avoid too many coincidences, in Figure Two we provide in clockwise fashion the OLS regression lines amongst the variables for government and external debt against economic growth in our panel of young South American democracies. Basically, there is a negative relationship between debt and economic activity, which suggests not only a negative economic relationship between debt and growth in the region, but also that growth determines debt1.

Figure 2: OLS regression lines, government debt, external debt, government and economic growth, South America, 1970-2007. Sources: IMF, World Bank and PWT …les.
OLS regression lines, government debt, external debt, government and economic growth

In essence, the above descriptive evidence, with all its caveats, suggest that the size of government and external debt, and growth have been moving in opposite directions, or alternatively that debt has decreased with faster economic activity. Alternatively speaking, the evidence suggests that those economies seem to be countercyclical after all. Above all, the tax-smoothing model is valid in the region in the sense that debt increased rather dramatically during the political and economic transitions that the region went through in the 1980s (the war period), however the economic recovery that followed (the peace period) has played an important role in reducing debt in those young democracies of South America (Barro (1979), Easterly (2001), and Hall and Sargent (2010)).

Final observations

Needless to say that South America has matured enormously from the time when a particular political party being elected would be an excuse for another junta to come into power. Moreover, this political maturity has been, to a certain extent, matched by an economic maturity, with better economic institutions and policies being implemented, not to mention that the quality of the civil service is much better than back in the 1980s.

However, perhaps the main lesson from the above analysis is the importance for a return to the basics in terms of understanding government and external debt, and the role and relevance of economic activity and prosperity in keeping debt under control. This is interesting in itself because the main policy implication coming from the analysis is about promoting a nondistortionary and nonpredatory environment that is conducive to economic activity, which somehow contrasts with some of the interventions and policies being recently implemented in countries like Argentina and Bolivia, which are more along the lines of not generating consistently faster economic activity.


Barro, R. "On the Determination of Public Debt." Journal of Political Economy, 1979, 87, pp. 240-71.

Bittencourt, Manoel. "Is Copacabana Still the ‘Little Princess of the Sea?’" CES-Ifo Forum, 2011, 12(1), pp. 11-16

Dollar, David, and Aart Kraay. "Growth is Good for the Poor." Journal of Economic Growth 2002, 7:195-225.

Easterly, William. "Growth implosions and debt explosions: do growth slowdowns cause public debt crises?" Contributions to Macroeconomics, 2001, 1:1.

Gavin, Michael and Roberto Perotti. "Fiscal Policy in Latin America."NBER Macroeconomics Annual, 1997, 12, pp. 11-72.

Hall, George, and Thomas Sargent. Interest rate risk and other determinants of post-WWII U.S. government debt/GDP dynamics. In NBER Working Paper Series, 2010.

* Department of Economics, University of Pretoria, Lynnwood Road, Pretoria 0002, RSA, e-mail:

1. 1For more on the estimation and on how we deal with issues such as endogeneity, please refer to the working paper version of this article

Share this