Cohen disputes the conventional argument explaining the African tragedy: the ‘ethnic’ diversity of the African nations explaining why they could not agree on an efficient set of institutions that might foster growth; and the degree of trade liberalization explaining why African economies failed to capitalize on world wealth to catch up with other nations. The problem with these interpretations is that they leave intact the Latin American tragedy. Accounting for ethnicity or trade liberalization always leaves Latin America’s growth rate 1.5% behind the performance of other countries.

 

• There is another difficulty with trade liberalization: when controlling directly for the degree of openness itself, conventionally measured as trade over GDP, the ratio appears to play no role in explaining growth (when account is taken of investment). One answer to this puzzle may be that openness is actually a poor indicator of the ability of a given country to import the goods it needs.

 

• Typically, a large country will always appear to be less open than a small one. The surprising feature reported here is that when openness is corrected for size, the corresponding measure appears to be (significantly) negatively correlated with growth! Splitting the sample in to those who have liberalized trade and those who haven’t, trade seems to hurt the latter, but not the former. Trade liberalization appears to be a means of dampening the harmful effects of trade on protected countries, rather than a means of raising growth in open economies.

The question to answer then becomes, through which channel is it that trade hurts the protected economies, rather than through which channel is it that trade fosters growth opportunities?

Cohen gives a twofold answer.

 

• First, exchange rate mismanagement is one key channel through which trade hurts protected economies. When the exchange rate is overvalued, the more open the economy, the worse its growth performance.

 

• The second dimension relates to debt crisis. Latin American economies have long been subject to the risk of debt crisis. Closed economies do not risk much by threatening to default, but, along with exchange rate mismanagement, external debt mismanagement does hurt countries open to trade.

By accounting for the role of the debt crisis influence, it is possible also to construct new solvency indicators. The World Bank has set two external debt benchmarks: a debt-to-export ratio above 220%; and a debt-to-GDP ratio above 80%. Both ratios have the merit of successfully tracking most of the debt crisis episodes. In retrospect, they appear to fit many characteristics of countries in debt trouble (discount on secondary market price, years of the first rescheduling...).

Beyond the debt-to-export and the debt-to-GDP ratios, he demonstrates that a debt-to-tax ratio also performs extremely well in predicting the risk of a debt crisis. The corresponding critical values above which this risk appears to have the largest negative effects on growth are: debt-to-export above 200%; debt-to-GDP above 50%; and debt-to-tax above 300%.

For the first two indicators these results are more conservative than the figures from the World Debt Tables. The third threshold is still tentative, but appears to be a very promising indicator.

‘Growth and External Debt:
A New Perspective on the African and Latin American Tragedies’
Daniel Cohen

Discussion Paper No. 1753

Available for £5/$8/€8 plus a postage and packaging cost of 50p/$1/€1 (UK or Europe) or £1/$2/€2 (Rest of World) from
CEPR, 90-98 Goswell Road, London EC1V 7RR, UK
Tel: (+ 44 20) 7878 2900 Fax: (+44 20) 7878 2999 Email: orders@cepr.org