pre-determined the finding of convergence discussed earlier in this chapter. A
useful finding from this chapter is that growth and stagnation occur in distinct
episodes. These episodes can be utilized as easily identifiable case studies and
the economic, political, institutional and policy conditions that accompany
these break points can then be examined as a case study.
For example, Hausman et al. (2004) find that the vast majority of growth
accelerations are unrelated to standard determinants such as political change or
economic reform. They find that only 14.5 per cent of growth accelerations are
associated with liberalization and 85.5 per cent are not. Jong-a-Pin and De
Haan (2011), using an index containing a number of features (presence of
marketing boards, a socialist economic regime, a large black-market premium
for foreign currency and tariff and non-tariff barriers to trade) to measure
‘liberalization’, find a significant link between changes in their index and
subsequent accelerations in growth. Various scholars have found that trade has
quite consistently expanded for African countries undergoing growth acceler-
ations (Hausman et al., 2004; Jones and Olken, 2008). However, this was
caused not by greater trade liberalization increasing the volume of imports and
exports,but by the expansion of trade (particularly through higher commodity
prices) after 1995 among the resource-rich countries.
Evidence linking specific policies to growth acceleration is also mixed but
does offer some clearer results. Exchange rate depreciations are commonly
found in the general literature to have been associated with growth accelera-
tions (Hausman et al., 2004). There is clear evidence to show that the extent of
exchange rate overvaluation declined sharply in Africa after the 1990s so the
timing does seem to be closely linked to growth accelerations. The real effec-
tive exchange rate in Africa, for example, halved between 1975 and 1994, and
from 1995 to 2005 (Arbache and Page, 2009:19).
Increased investment is also commonly associated with growth episodes in
both developing and developed countries (Hausman et al.,2004; Jones and
Olken,2008),with the exception of Africa. Here, increased growth after 2000
was accompanied by only marginally increased investment rates: from 15.25
per cent share of GDP between 1975 and 1994 to 16.69 per cent between 1995
and 2005 (Arbache and Page, 2009:19). This remains a long way below the 30
per cent+ shares of investment experienced by East Asian countries during
their economic booms of the 1970s and 1980s, and by China after 1980 and
India after 2003. Africa joined the wave of global democratization in the 1990s
and beyond. Although political and popular sentiment has often linked this to
improved macroeconomic outcomes after 2000, scholars have found find that
autocracy has a strong and positive relation to growth episodes (Hausman et
al.,2004), that autocracies are 70 per cent more likely to experience growth
accelerations (Jones and Olken,2008) and even that democracy is a robust
predictor of growth declines (Cuberes and Jerzmanowski, 2009). Jong-a-Pin
and De Haan (2011) are unusual in finding that a move towards democratiza-
tion is often associated with growth episodes but that this effect is only tempo-
rary and declines the longer the democracy has been in place.
54 Sources of Growth in the Modern World Economy since 1950