The IMANI think tank has recently published a series of articles opposing aspects of the proposal by the flag-bearer of the opposition New Patriotic Party (NPP) to offer “free” senior high school (SHS) education to all Ghanaians. Specifically, IMANI contends that implementing the policy would cost the public purse nearly three times more money than has been budgeted by the NPP. In a meta-critique of IMANI’s reactions, Dr. Nii Moi Thompson takes IMANI to task for what he calls “a disastrous exercise in disinformation and fabrication of facts” (see “IMANI and the Great Education Debate: Separating Fact from Fiction,” Ghanaweb, April 19, 2012).
Dr. Thompson’s beef appears to revolve around three issues. First, IMANI appeared to credit the NPP with the Free SHS policy, instead of the CPP, which was the first to propose the policy in its 2008 Manifesto. Second, IMANI engaged in hasty and faulty data extrapolation and analysis that overestimated the cost burden on the public purse ostensibly to justify its ideological opposition to the policy. Third, in all its publications IMANI appeared obsessed with cost, while ignoring questions of equity and social justice. Even with 40% of Ghana’s income, Dr. Thompson contends, Uganda is already implementing a similar policy.
In the following paragraphs I use the contest between Dr. Thompson and IMANI to explore questions about ideology and the use of data in development debates as they relate to this particular policy. To begin with, most readers would probably agree that whether the NPP or the CPP was the first to propose the policy is quite irrelevant to the debate, not least because the CPP, as currently constituted, is unfortunately in no position to win a presidential race, let alone get the chance to implement its “proprietary” policies.
The two other issues are more fundamental to the debate. In their own rebuttal to Dr. Thompson’s meta-critique, IMANI admit possible errors, even if indirectly, but attribute any errors to data sources rather than to faulty analysis on their part (see “Nii Moi Thompson and the Cheerleaders of Free SHS,” Ghanaweb, April 24, 2012). To be sure, most data you can find, whether sourced from local or international agencies, are abstractions that do not always correspond to the effective empirical reality. Still, a researcher’s data selection criterion and analysis can reduce accuracy even further.
Putting Ideology Ahead:
The problem with IMANI’s rebuttal is that the acknowledgment of possible errors does not encourage self-criticism of the appropriateness of their methods. One is therefore left to feel that IMANI approached the analysis with a pre-determined conclusion, so that insofar as this conclusion was reached questions about the accuracy of the data became immaterial. It doesn’t also take much to discover the anti-state ideological roots of IMANI’s opposition to the policy, as Dr. Thompson correctly observed with references to IMANI’s use of words like “clunky state”.
Given that the historical evidence overwhelmingly supports the use of state power to promote both economic development and social justice in poor countries, IMANI’ anti-state agenda is problematic, not least because it often affects the way they problematise and explain serious policy issues. For example, rather than propose ways to build the capacity of state institutions, they engage in needless vilification of state institutions. Given the option between state intervention and schools under trees, they would choose the latter because the state is an avoidable, not a necessary, evil.


Manipulating Statistics:
In an attempt to pay back Dr. Thompson in his own coin, IMANI faulted his claims on the size of Uganda’s income relative to Ghana’s. In IMANI’s own words: “Uganda does not have ‘40% of Ghana’s income.’ This is patently false. Uganda’s current GDP is $17.1 billion according to the World Bank, compared to Ghana’s $31 billion. That is at least 55% of Ghana's GDP.” IMANI then proceeded to refer to this supposed error as a “flagrant massacring of data and facts” that “does not do much credit to [Dr. Thompson’s] cause, whatever it is.”
Unfortunately, it is IMANI who are undermining their own credibility and cause by always rushing out with hasty analysis to beguile unsuspecting readers. If they had taken their time, they would have been prompted to check the kind of data Dr. Thompson selected for his comparison on incomes and to ask which approach was more appropriate in the specific context of the debate. Incidentally, both IMANI and Dr. Thomspon sourced their data from the same World Bank database. But whereas IMANI used the nominal size of GDP as at December 2010, Thompson used GDP per capita for the same period.
Uganda's GDP in current prices in 2010 was US$17.01 billion (US$17,010,765,766.70), roughly 54.34% of Ghana’s US$31.3 billion (US$31,305,891,329.22). Now, notice that the figure for Uganda’s GDP is different from that given by IMANI (US$17.1bn), which allows them to state a slightly higher percentage for Uganda. This could be an error but it also indicates how data can be manipulated to suit particular ends.
From the same World Bank database, Uganda’s GDP per capita in 2010 was US$508.95 in current prices, compared to Ghana's US$1,283.46. This means that the average income per person in Uganda was about 39.65%, roughly 40%, of the average income per person in Ghana. Obviously, this is the dataset used by Dr. Thompson.
Of course, there is a problem with using current dollar prices for international comparisons. The standard practice is to use purchasing power parity (PPP). But since both IMANI and Dr. Thompson have used current dollar prices, we can drop that argument for now and proceed to ask which of the above datasets should be used in the specific context of the debate. It seems to me that Dr. Thompson was right in using GDP per capita since this debate is about: (i) whether Ghanaian households can afford SHS education for their wards and (ii) if they cannot, whether the state should intervene and by how much.
GDP per capita is a better measure of how much of the national income goes to individuals. In other words, if income was evenly distributed, the average Ghanaian should be better off than the Ugandan. By using GDP per capita analysis, we’re also able to address concerns about both equity and cost. Let’s address the two issues one at a time.
Equity and Social Justice:
On equity, there are two ways to interpret the data. One is that it's OK for Uganda to offer free education to its citizens, since income per head in Uganda is comparatively lower than in Ghana. But this interpretation would be sound only if incomes were evenly distributed in both countries. The World Bank Gini index, which measures “the extent to which the distribution of income or consumption expenditure among individuals or households within an economy deviates from a perfectly equal distribution,” gives us a fair sense of the distribution of income in both countries. The Gini index is 44% for Uganda (2009 data) and 43% for Ghana (2006).
The average Ghanaian is not necessarily better off than the Ugandan because they don't obtain a fair share of domestic output. In principle, this affirms Dr. Thompson's argument: if Uganda is able to offer free education, Ghana certainly can, because it has more income that it is not sharing equitably.
Cost Implications:
At the same time, however, the question arises whether those who obtain a disproportionate share of national income should compete with the poor for free education. This is where IMANI's argument for means testing must be taken seriously. First, a free SHS policy that doesn’t discriminate between rich and poor households may actually end up giving more to the rich by way of freebies that they don't need or can do without. This would reinforce inequality since it makes available even greater resources to the rich.
Secondly, by so doing the policy would strain the public purse more heavily than if the rich had been made to contribute. Thirdly, and arising from the foregone, the policy would require shifting resources away from other important areas of public finance, such as health and physical infrastructure. Here, it is important to remind ourselves that in spite of the transition into lower middle income Ghana is still a very poor country that needs to continue raising investment spending to create the environment for sustained cross-sectoral growth.
Our new status as an oil economy makes this imperative to avoid the oil curse. It is equally important for avoiding a slide back into low-income status. Moving up into upper middle income isn’t automatic. The income classifications are periodically reviewed as the richest countries become even richer. By the income classifications in the 1950s and early 1960s, Ghana was a lower middle income country. We slid back by being stagnant or growing only marginally for a long time.
To sustain growth and diversify the economy, we need more than human capital. We need skilled human capital to be globally competitive. It is not enough to expand access to education. We must invest in quality. Similarly, skilled human capital is not enough. Adequate physical infrastructure would be needed to facilitate investment and improve labour productivity. In spite of the abundance of technical skills, India for a long time suffered the infamous “Hindi growth rate” due partly to the absence of adequate infrastructure.
The flag-bearer of the NPP Nana Akuffo Addo has been quoted as saying that he would not mind spending all the oil revenues on education. Given how oil revenues have been squandered by our neighbours in the Gulf of Guinea, one can understand why he would say this. But this is not sound policy either. By encouraging rich households to contribute toward the cost of education while targeting the poor with scholarships, we can achieve many of the above goals with our limited resources. By approaching the subject with open minds, we can arrive at a fruitful common ground somewhere between the cost advocates and the social justice advocates.

Kobina Idun-Arkhurst