There is a long history of academic work on whether inequality affects growth. Boushey and Price (2014) provide an accessible survey of the literature in which they conclude that, while theory points to “numerous possible mechanisms” with “conflicting outcomes,” the evidence is now clear: “As new data become available and better data analysis methods are applied … studies that look at the longer-term effects find that inequality adversely affects growth.”
Previous studies, however, have neglected a key feature of the growth process, which is particularly important in developing countries: its lack of persistence. Average incomes do not typically grow steadily for decades. Rather, periods of rapid growth are punctuated by collapses and sometimes stagnation—the hills, valleys, and plateaus of growth. Relating inequality to medium-run average growth may thus miss the point. The more relevant issue for many countries is whether inequality is related to these sharp growth breaks.
Figure 3.1 shows that the evolution of average incomes in advanced countries, such as the United States and the United Kingdom, is consistent with the common perception that development is similar to climbing a hill: more-or-less steady increases in income, with small bumps along the way. If this were the universal pattern, the most interesting question would be how to explain why some countries grow faster than others over long periods. But the figure also shows that in emerging markets and developing countries—such as Brazil, Cameroon, Chile, and Jordan—there is a variety of experiences. Looking at such pictures, Pritchett (2000) and other authors have been struck that an understanding of growth must involve explaining why some countries are able to keep growing for long periods of time, whereas others see growth downbreaks after just a few years, followed by stagnation or decay.
FIGURE 3.1: Growth Patterns in Advanced and Developing Countries
The common perception of development as a steady process holds true in some advanced economies (as depicted for the United Kingdom and the United States), but developing countries experience a bumpier process.
Note: Vertical dashed lines represent statistically significant growth downbreaks; solid lines represent upbreaks. Real GDP per capita is measured on a logarithmic scale, so a straight line implies a constant growth rate.
Source: Berg and Ostry (2017).
The question of how to sustain growth is particularly interesting for two reasons. First, igniting growth has proven much less challenging than sustaining growth (Hausmann, Pritchett, and Rodrik 2005). That is, even the poorest of countries or countries with poor economic institutions have managed to get growth going for several years from time to time. Where growth laggards differ from their more successful peers is in their inability to sustain growth for long periods of time. Second, since the 2000s, a large number of countries have experienced sustained growth. This is particularly the case in sub-Saharan Africa, where many countries had a takeoff in the mid-1990s. Is this growth likely to persist, and how can it be kept going?
We focus therefore on the duration of “growth spells”—starting when growth takes off (the “upbreak”) and ending when growth slacks (the “downbreak”). We then try to explain how this duration depends on various policies and country characteristics, including inequality. Because the goal is to examine trends, not temporary events such as recoveries from recessions or booms, we look only at growth spells that have lasted at least eight years.
As we started to pull together the data, it became evident that both upbreaks and downbreaks were indeed quite common. As figure 3.2 shows, upbreaks tend to be fairly spread out across regions and have occurred in every decade. A key message from the data is thus that the initiation of growth is not the hard part of trying to achieve a long-run rise in per capita incomes. Rather, the real problem is the inability to sustain growth over long periods. For example, almost all growth spells in advanced countries and in emerging Asia last at least 10 years or more, but only about two-thirds of Latin American and African spells do (figure 3.1).
FIGURE 3.2: Growth Breaks by Region and Decade
Upbreaks and downbreaks are quite common and tend to be fairly evenly spread out across regions.
1/ Percentage point change in real per capita GDP growth before and after the break.
2/ Includes Japan, Korea, Singapore, Hong Kong SAR, and Taiwan Province of China.
3/ Middle East, North Africa, Cyprus, Turkey, and Caribbean countries.
Note: A growth break is a statistically significant change in the per capita real GDP growth rate that persists for at least eight years.
Source: Based on Berg and Ostry (2017).
Another salient feature of the data is the rate of growth when the country is in a spell versus the aftermath of a spell. As table 3.1 shows, all regions’ spells involve fairly fast growth, with those in Africa actually the most rapid. In contrast, there are big differences across country groupings following the end of spells. African spells have tended to end in deep collapses, where as soft landings have tended to follow the end of growth spells in advanced countries and Asia.
TABLE 3.1: Characteristics of Growth Spells
1/ Real per capita GDP growth, in percentage points.
2/ Includes Japan, Korea, Singapore, Hong Kong SAR, and Taiwan Province of China.
3/ Middle East, North Africa, Cyprus, Turkey, and Caribbean countries.
Note: A growth spell is a period between a growth upbreak and a growth downbreak, as long as per capita real growth is above 2 percent during the spell and falls to below 2 percent after the downbreak. Breaks are at least eight years apart.
Source: Berg, Ostry, and Zettelmeyer (2012).
Having identified the growth spells over the last half century, we then turned to the question of how the duration of these spells relates to differences in inequality. Figure 3.3 illustrates the relationship between the duration of growth spells and the Gini measure of income inequality for a sample of countries. There is clearly a promising pattern here: more inequality seems associated with less sustained growth.
FIGURE 3.3: Duration of Growth Spells and Inequality
More inequality is associated with less sustained growth.
Note: This figure includes spells that end in-sample (completed spells) only, because the length of incomplete spells is unknown. For this figure, minimum spell length is eight years. Countries include Belgium, Chile, Denmark, Dominican Republic, Ecuador, Egypt, France, Greece, Guatemala, Indonesia, Ireland, Jamaica, Japan, Jordan, Malawi, Nigeria, Pakistan, Philippines, Portugal, Republic of Korea, Senegal, Syrian Arab Republic, Taiwan PoC, Thailand, and Tunisia.
Source: Berg and Ostry (2017); updated.
This figure is only the starting point of our analysis, however. Many other factors could determine the duration of a country’s growth spells: the quality of its economic and political institutions, openness to trade, macroeconomic stability, and the skills of its labor force (human capital accumulation, in the jargon). Our next step therefore is to consider how some of these other factors affected the duration of growth spells, and to explore whether inequality still played a role once the impact of these other factors was accounted for.
Our approach borrows from the medical literature that aims to gauge, for example, how long someone might be expected to live conditional on certain factors (whether they smoke, their weight, their gender, their age, etc.). In our context, the probability that a growth spell will end depends on its current length and various “hazards” to growth. The analysis distinguishes between initial conditions at the onset of the spell and changes during a growth spell.
Unfortunately, there are not enough data to test at once all the theories about what could affect growth spells. There are simply too few spells and too many candidate explanations to disentangle everything. Moreover, many of the candidate determinants of spell duration could themselves be correlated with one another. Hence, we did some preliminary analysis to find a set of candidates that appeared to show promise, ones that ended up as important in most samples and specifications. On this basis, we ended up with six factors that played a significant role in explaining the duration of growth spells: (1) political institutions, (2) openness to trade, (3) exchange rate competitiveness, (4) foreign direct investment (FDI), (5) reliance on external debt, and (6) inequality.
The first five of these factors are not much of a surprise given what is already known about the determinants of growth. Each of these factors has been mentioned as being potentially important in that vast literature. Many have argued that political institutions that constrain the executive and secure accountability help growth (Sokoloff and Engerman 2000). There is also evidence that trade liberalization helps growth by increasing market size, promoting competition, and making it easier to transmit know-how (Alesina, Spolaore, and Wacziarg 2005). An overvalued exchange rate can undermine growth by making the manufacturing sector less competitive (Rodrik 2008). And while the growth benefits of foreign capital flows are difficult to establish, as we discuss in later chapters, some flows such as FDI do seem to help growth, as does lower reliance on external debt (Dell’Ariccia et al. 2008).
We find these factors to be also important for the duration of growth spells. The duration is higher in countries that have better political institutions, are more open to trade, maintain a competitive exchange rate, attract more foreign direct investment, and have lower external debt.
Our results show that inequality belongs as a sixth factor in this pantheon of determinants of the duration of growth spells: Countries with a more equal income distribution have longer spells. Inequality retains its statistical and economic significance despite the inclusion of the other factors discussed above. This suggests that inequality matters in and of itself and is not just picking up the effects of other factors. Inequality also preserves its significance more systematically across different samples and definitions of growth spells than the other factors. It is thus a more robust predictor of growth duration than many variables widely understood to be central to growth.
Overall, the results of the analysis support some interpretations of why East Asian economies grew rapidly between 1965 and 1990—a development that has come to be called “the East Asian Miracle” (World Bank 1993). Growth is most enduring in countries that are open to trade, attract foreign direct investment, maintain low external debt, and have relatively equal income distribution. Given this, it is worth noting that our overall results hold up even when Asia is excluded from the sample.
Figure 3.4 provides a view of the importance of each variable. It reports the increase in expected spell duration for a given increase in the variable in question, keeping other factors constant. There is a large and significant association between income inequality and growth duration. A 10-percentage-point decrease in inequality—the sort of improvement that a number of countries have observed during their spells—increases the expected length of a growth spell by 50 percent.
FIGURE 3.4: The Effect of Increase of Different Factors on Growth Spell Duration
Many factors increase the duration of growth spells but a more equal income distribution is one of the more important factors.
Note: For each variable, the height of the figure shows the percentage increase in spell duration resulting from an increase in that variable from the fiftieth to the sixtieth percentile, with other variables at the fiftieth percentile. For trade, the figure shows the benefits of having an open instead of a closed regime, using the Wacziarg and Welch (2008) dichotomous variable. For political institutions, the figure shows the effects of a move from a rating of 1 (autocracy) to 0.
Source: Berg and Ostry (2017).
The channels through which income inequality affects the duration of growth can be both economic and political. One economic channel—which economists put under the category of credit market imperfections—is that poor people do not have the means to finance a good education or afford quality health care. Hence, they are unable to build up their human capital and this makes it difficult to sustain the economy’s growth. This echoes the arguments in Wilkinson and Pickett (2011) that more unequal countries suffer from relatively poor social and health indicators, such as increased illness and anxiety.
Income inequality may also increase the risk of political instability. In unequal societies, there may be greater tussles between the rich elites and the rest for political power, and the resulting uncertainty could reduce incentives to invest and hence impair growth (Alesina and Rodrik 1994). Beyond this, the ability of unequal societies to respond effectively to adversity may be lower because it may be more difficult to rally everyone to a common cause (Rodrik 1999, b). This greater difficulty reflects the fact that, in good times, the “have-nots” were left behind; thus, they see little point in supporting the pain of societal adjustment in tough times when the prospects of future gains for them are slim.
Our statistical results on inequality and the duration of growth, and our conjectures about the channels through which these results come about, are reflected in the political and economic narratives of many countries. Table 3.2 lists the six cases where our statistical analysis indicated the risk of the spell ending was the highest among all the countries in our sample. In the economic histories of these cases, we often found mention of the role of income distribution in bringing the spells to an end.
Colombia experienced a spell end in 1978. Our statistical model predicts that Colombia’s growth spell was indeed fragile—with a risk of ending sixty-six times higher than the average over all the spells in our sample. This higher risk can be decomposed into the six determinants included in the duration results: Colombia’s high Gini (53 vs. the sample average of 38) accounts for most of the higher risk. While the spark was a crackdown on drug cartels, beginning a long civil conflict, Cardenas (2007) notes the role of “high levels of inequality and poverty and the weak presence of the state.” Guatemala was in a state of civil war from 1960 until 1996. In the words of Thorp et al. (2006): “By the late 1970s, Guatemala had entered a stage of polarization and radicalization of social organizations (trade unions, peasant organizations).” Carbonnier (2002) further notes that “attempts to cut subsidies and raise transportation prices repeatedly spurred violent clashes in the streets of Guatemala City.” The prediction of the model is that the risk of Guatemala’s spell ending was indeed about fifty-five times higher than the average country during 1974–1979, with higher-than-average income inequality being one of the main factors driving the result.
The ends of spells in Cameroon, Nigeria, and Ecuador demonstrate how income distribution can interact with adverse external developments. Lewis (2007) notes for Nigeria that highly volatile politics, lack of social cohesion, and external shocks drove bursts of economic volatility: “In Nigeria, ethnic and regional competition has hampered the formation of a stable growth coalition between the state and private producers. Political elites have turned instead to populist strategies and diffuse rent distribution among a fragmented and polarized business class. The populist option proved short-lived when oil revenues dwindled, while the residual rentier alliances were unstable, resulting in economic stagnation and disarray.” In Cameroon and Ecuador, oil wealth in the 1970s initially financed large increases in the public sector, particularly in the wage bill, which proved very difficult to cut when oil prices fell. “Although these measures were necessary to rescue [Cameroon] from further economic crisis, they were very unpopular because they least affected the political elite and those in the upper echelon of government, whose privileges remained intact” (Mbaku and Takougang 2003; see also Jácome et al. 1998 and Aerts et al. 2000). In all three countries, the model’s hazard ratio was very high (ranging from more than one hundred times higher than normal for Cameroon to twenty-nine times higher for Nigeria) and high inequality and autocracy levels as well as low levels of FDI, played important roles in bringing the growth spell to an end according to our statistical results.
The model attributes the (high) risk of Panama’s spell ending mainly to rising external debt, along with inequality. Indeed, Panama’s military dictatorship preserved power through the 1970s increasingly by borrowing externally to support transfers to government workers (Ropp 1992). The global crisis of the early 1980s thus hit Panama hard. This pattern is consistent with the argument in Berg and Sachs (1988) that countries that suffered most from the debt crisis of the 1980s may have been those that used (unsustainable) foreign borrowing to bridge societal conflict.
TABLE 3.2: The Ends of Six Growth Spells
Note: The hazard ratio is the predicted probability that the spell would end during the five years prior to its actual end, as a ratio to the predicted probability of a spell ending for the average observation in the entire sample. Thus, a hazard ratio of 1 implies no unusual risk that the spell will end. See Berg, Ostry, and Zettelmeyer (2012) and the technical appendix to this book for details.
Source: Berg and Ostry (2017).
The variety and complexity of the channels are evident in these examples. The timing of crises seems to reflect an interaction of underlying vulnerabilities, including income distribution. Clearly, ethnic fractionalization plays a role in some cases too. The statistical evidence shows that inequality is an underlying feature that makes it more likely that a number of these factors come together to bring a growth spell to an end.
