Complexity and ambiguity – perhaps it is in these two aspects where what I call ‘ambivalence of numbers’ lay. I refer to this term as the sense of mixed-feelings about the information depicted by quantitative data sets that can arise amongst practitioners, researchers and other actors in the arena of international development. Numbers can provide clear-cut, broad images of a phenomenon, which can be a useful tool to dig deeper into the intricate web of its underlying forces. This is by no means a new idea. In fact, a general consensus has been established around the use of mixed-methods for data accuracy in social inquiry and its ethical representations (Dogra, 2012; Greene, 2008; Scheyvens, 2014). However, this does not seem to be practiced sufficiently through the mainstream media discourse on international development. The recent publication by Anderson (2014) entitled 'Rapidly developing states at risk of rising corruption', as published in The Guardian, is one such example. The following will provide a brief review of the article by critically assessing the methodological limitation of the article, especially in the context of Sub-Saharan Africa (SSA). It will also shed light on challenges around the use of other alternative instruments, such as Ibrahim Index.
GDP Annual Growth 4% ≥ Raising Corruption?
Drawing on Corruption Perceptions Index (CPI) composed by the Transparent International [TI] (2014), Anderson (2014) reports that developing countries whose annual GDP growth is over 4% are likely to fall deeper into the state of corruption – characterized predominantly by the propensity to bribery and fraud. Notably, the article acknowledges the methodological limitations of CPI. In particular, it shows that it has become the mainstream instrument in the space of corruption measurement “without providing hard evidence” around the ranking (Anderson, 2014: para. 8). It also points out that the CPI may not capture every dimension of corruption yet still remain a useful tool to aid the wide range of general public audience. This includes practitioners and academics to ‘imagine’ what is going on in the arena of public (administrative-political) corruption.
Danger of Quantitative Driven Imagination
This notion of ‘imagination’ driven by quantitative figures and its reproduction mechanism, however, need critical attention. On the one hand, the article provides a limited and misleading image of public sector corruption due to its failure to provide alternative measurement to complement the ambivalence of figures being discussed. For instance, the article places a strong emphasis on Somalia as a country that is most prone to bribery (along with North Korea), given its lowest score and deepest fall in scores amongst other SSA countries and different regions. However, it does so without explicitly mentioning that the availability of data on the country’s GDP annual growth is significantly limited. In fact, the last updated official figure of -1.6% was collected in 1990 (World Bank, 2014a). Thus, ultimately reinforcing the image of Somalia as a poor and corrupt African country, as though the country’s state of corruption is merely resulted from the high GDP annual growth rate. Given that such a manipulative practice is done by mainstream media, like The Guardian, it raises a critical question of to what extent should such mainstream discourse be just.
Moreover, the article fails to explain what causes the ‘depth’ of corruption in the context of rapid growth without development. For instance, the article states that Sub-Saharan African countries ranked lowest in the CPI with the majority being scored under 50, which “depicts endemic corruption” (Anderson, 2014: para. 15). It appears very selective about the type of data being presented while being unclear on the definition of ‘depth’ of corruption. As such, no evidence is provided around the cross-country variations amongst the GDP annual growth rate . Nor does it show other social-human development indicators – let alone any qualitative data or ‘soft evidence’. The only ‘hard evidence’ that it provides is the significant illicit financial flaws of USD 60 billion per annum as a representation of high level of corruption as a region. Drawing from the data set provided by TI (2014) and WB (2014a), the number of SSA countries whose CPI fell by more than -1 and whose GDP annual growth rate was over 4% appears to be 12 out of 47 listed SSA countries in the CPI index (see Table 1). This certainly raises the question of generalizability around the article’s core claim as well as the need for further inquiry on underlying factors that perpetuates the ‘epidemic’ nature of deep public corruption in the rest of the countries in the region.
On the other hand, such ‘ambivalence of numbers’ can also be found when applying alternative measurements, such as Ibrahim Index (II). In an attempt to understand the multi-dimensional nature of corruption – which is ignored by Anderson (2014) for its limited focus on CPI, the application of such an index can be a potential response. One of the useful features about the II is that it incorporates a wider range of underlying indicators that go beyond the sphere of administrative-political corruption, such as human development, including Gini Index (GI) numbers (Mo Ibrahim Foundation, 2014 and Rotberg, 2009). Certainly, it has been claimed that the state of corruption has a 'degenerative' impact on income distribution in SSA, regardless of the rate of economic growth (Gyimah-Brempong, 2003; Gyimah-Brempong & De Camacho, 2006). When running a simple regression test between II and GI of the concerned 47 SSA countries with the filter of GDP annual growth rate of over 4% based on Anderson's (2014) claim, however, only a moderate, positive correlation with relatively a weak association is demonstrated with r = 0.61 and r2 = 0.38. This in turn suggests that there are limitations in compatibility of II and GI in this context. It also implies methodological limitations of GI, such as inconsistency in data availability or sample size (WB, 2014b).
Numbers as Ambivalent Diagnostic Tool
As discussed above, the ambivalence of numbers should always be revisited when evaluating the validity of quantitative analysis, especially if it is provided through powerful knowledge dissemination mechanisms. As asserted by Rotberg (2009: 113), quantitative indexes, such as II and CPI, should be used as “a diagnostic tool for civil society, donors and governments” so that not only the state of administrative-political performance but also the lives of people in Africa are enhanced. Further, gender dimension remains excluded from the existent measurements of corruption due to its ‘soft’ and ‘hard-to-measure’ natures in comparison to the conventional ones (Seguino & Were, 2014). An enhanced understanding of supply-demand impacts that gender equality can have on macro-performance is sought to be critical particularly in SSA which contains “seven out of the ten most gender-unequal countries in the world” (ibid.: 20). When pursuing such an approach, ambiguity and complexity that arise in the development process should receive careful attention. Perhaps, this is where the mainstream media, like The Guardian, should play a critical role.
- Anderson, M. (2014). Rapidly developing states at risk of rising corruption, The Guardian, 4 December, 33.
- Dogra, N. (2012). Representations of global poverty: Aid, development and international NGOs. London: I.B. Tauris.
- Greene, J.C. (2008). Is mixed methods social inquiry a distinctive methodology? Journal of Mixed Methods Research, 2 (1), 7-22.
- Gyimah-Brempong, K. (2003). Corruption, economic growth, and income inequality in Africa. Economics of Governance, 3, 183-209.
- Gyimah-Brempong, K, & De Camacho, S.M. (2006). Corruption, growth, and income distribution: Are there regional differences? Economics of Governance, 7, 245-269.
- Mo Ibrahim Foundation. (2014). 2014 Ibrahim Index of African governance: Summary report. Available at: www.moibrahimfoundation.org [Accessed 01/03/2015]
- Rotberg, R.I. (2009). Governance and leadership in Africa: Measures, methods and results. Journal of International Affairs, 62 (2), 113-126.
- Scheyvens, R. (2014). Development fieldwork: A practical guide. London: SAGE Publications Ltd.
- Seguino, S., & Were, M. (2014). Gender, development and economic growth in Sub-Saharan Africa. Journal of African Economics, 23 (1), i18-i61.
- Transparency International. (2014). Corruption perceptions index 2014: Results. Available at: www.transparency.org [Accessed 01/03/2015]
- World Bank. (2014a). GDP growth (annual %). In World Development Indicators. Available at: http://data.worldbank.org/ [Accessed 01/03/2015]
- World Bank. (2014b). Gini index (World Bank estimate). In World Development Indicators. Available at: http://data.worldbank.org/ [Accessed 01/03/2015]