Good Growth and Good Governance in Africa: An Experts Forum

By Jeffrey Herbst, Tim Kelsall, Goran Hyden, and Nicolas van de Walle

Richard Joseph

In 1993, IMF Director Michel Camdessus called sub-Saharan Africa “a sinking continent.” Just three years later, however, he said that the signs of an economic recovery were evident. 1 In his state visit to Ghana in July 2009, President Barack Obama saluted the striking turnaround in the political economies of Africa. He memorably gave the main reason for this historic advance “Development depends on good governance. That is the ingredient which has been missing in far too many places, for far too long. It is the change that can unlock Africa’s potential.”

In a decade and a half, Africa had gone from being a sinking to a “rising” continent, a change not deterred by the 2008-2010 global recession. Is this experience attributable to much improved governance in sub-Saharan Africa? If that is the case, what are the mechanisms by which better governance unlocked economic growth? Further, it can be asked, how good must this governance be? As major economists returned to the study of Africa, voices emerged among them questioning a new “orthodoxy”. It was difficult, they contended, to connect the “good governance agenda” of democratic institutions, civil society, and human rights with the East Asian models of rapid transformative growth. 2 Notable among the critics was Mushtaq Khan of the University of London. He rejected the notion that good governance, broadly understood, was a precondition for economic growth. Instead, Khan argued, what was needed were “growth-enhancing governance capabilities.”3

Students of Africa were intrigued but puzzled by such contentions. Just what were these capabilities in the African context? Equally provocative was a theoretical framework, “developmental patrimonialism”, advanced by a program of Britain’s Overseas Development Institute (ODI) to characterize the authoritarian capitalism of the Ethiopian and Rwandan governments. 4 In response, it was pointed out that African countries with quite different political systems had also experienced sustained growth. In brief, clearly lacking was an agreed theoretical framework, backed by statistical data, to account for significant changes in African political economies.

Into this quandary step Pierre Englebert and Gailyn Portelance of Pomona College, USA. They suggested in their AfricaPlus essay that certain minimal advances in economic governance by African “developers” gave a “signal” to foreign investors. 5 Such a signal was missing in the case of “laggard” states. This minimalist argument appears consistent with Khan’s contentions but falls well short of the developmental patrimonial model.

Four leading scholars of Africa’s political economy accepted the invitation of AfricaPlus to respond to Englebert and Portelance’s arguments and their key hypothesis. Other scholars will be invited to join this important debate. Englebert and Portelance acknowledge the further empirical, statistical, and theoretical work that is needed. A better understanding of the economic upswing in Africa will be helpful in the pursuit of transformative growth, the much-awaited next stage in economic progress.6 Such growth—as would be reflected in diversified exports, revolutionized agriculture, greatly improved infrastructure, and employment-generating industrial and service sectors—is a concern in early 21st century Africa. While many rivers have been crossed in the past quarter-century, a mighty one remains to be forded.


Jeffrey Herbst

Englebert and Portelance have entered an important debate about the relationship between governance and growth in sub-Saharan Africa. Their “modest” (to use their term) hypothesis that relatively small changes in governance can produce signaling effects for foreign investment is at once intriguing and nuanced and needs, as they note, to be explored at greater length. Their investigation also demonstrates why a granular knowledge of African conditions offers much to general social science explorations. They can use the diversity of still-comparable African countries to develop testable ideas that have obvious import elsewhere.

Perhaps the most important contribution of the paper is to acknowledge the messy and inchoate nature of reform in Africa but still try to tie these efforts to results. As the authors note, while the trend in Africa has been toward greater freedom and economic reform, these efforts have been episodic and have not resulted in many institutionalized democracies or well-functioning economies. Not surprisingly given how difficult structural reform can be and the limited duration of these efforts, the African governance improvements are often difficult to describe. Thus, multiparty elections are held almost uniformly across the continent but vary in quality (from outright frauds to free and fair); democratic institutions have been established but struggle for funding and legitimacy; and corruption continues to be a significant problem even though some of the most pernicious economic distortions that promoted rent-seeking (e.g., the administrative allocation of foreign currency) have been eliminated.

Herbst pullquote

Thus, the argument that even modest changes in governance might have important economic effects is especially pertinent. Indeed, these changes can have an outsized impact on countries that have for so long suffered decline and where many may have given up hope for any change in state actions. In such a context, a change in trajectory of state performance, even if the slope is not that impressive, will be notable to all economic agents.

The major challenge for the authors as they move forward is to move beyond the foreign investment indicator. If governance changes are signaling events, surely the most important audiences are domestic. Peasant farmers, traders, and businesses of all types must be convinced that governance is improving if they are to invest. These audiences will be the most important for an economic turnaround. Indeed, it is an old convention that foreigners will only invest when they see that domestic agents—who inevitably know more about the country that outside firms—have enough confidence in the future stability of the country to risk their own capital.

It may be that the authors chose foreign investment as the major indicator because they believe that the modest governance reforms that the developing countries have adopted would not have any impact on potential new “winners”, and that regimes would try to retain exactly the same portfolio of supporters. However, this is probably too precise a calculation for most leaders and the real story will change from country to country. Indeed, leaders can probably make the kind of relatively minor governance changes that the authors detail without challenging their fundamental political calculus. A study which looks at signaling to both domestic and foreign actors would be particularly interesting and important. Overall, the promise of this study is a much more granular understanding of how reform happens and who wins and loses. Completing such a work would be a significant contribution to the understanding of African politics.

herbst

Jeffrey Herbst is President of Colgate University. He writes extensively on African politics, particularly on governance and development.

 

 


Tim Kelsall

Englebert and Portelance have written a well-researched and highly suggestive article that calls into question two competing and currently fashionable explanations for Africa’s recently revived growth experience. On the one hand, it is claimed that growth is linked to improved ‘democratic governance’ while, on the other, the reason given is a type of ‘developmental patrimonialism’. Based on a comparison of economic governance data for samples of African ‘developers’ and ‘laggards’, they find that governance is indeed better in developers, but that it remains quite poor. They also find that the difference in governance between these two groups has remained constant throughout most of the period during which growth has been diverging, with the origins of the difference traceable to a period in the early 1990s when the trend lines for the two groups crossed.

Scrutinizing these results, they are intrigued by the fact that ‘baby steps’ in governance appear to have a big impact on growth—a phenomenon for which other theories fail to account. In explanation, they offer the tentative hypothesis that, around the early 1990s, the ‘developers’ made a limited number of politically affordable governance changes that signaled their attractiveness as destinations for foreign direct investment. These signals succeeded in attracting disproportionate amounts of FDI, boosting growth, while leaving the neo-patrimonial logic of domestic politics largely unchanged.

Screenshot 2015-02-18 17.59.08

At a general level this argument seems plausible: it is compatible with what we know about the scale of governance changes, the persistence of neo-patrimonial practices in much of the continent, and what we know about the ‘herd mentality’ of investors. It is certainly convincing for Tanzania, the case I know best. At the same time, I am not sure this hypothesis invalidates the other competing perspectives. For example, when it comes to the ‘democratic governance’ explanation, it is perfectly possible that ‘developers’, with a view to earning popular legitimacy, use some of the changes to signal to their electorates the birth of a new order, with increases in FDI a positive side-effect. Thus democratization may still play a role in the explanation, if a reduced one, and via a less direct route.

When it comes to the ‘developmental patrimonialism’ thesis, Englebert and Portelance state that it is more persuasive for the Ethiopian and Rwandan cases than for other ‘developers’. As one of the authors of that thesis, I agree. The argument of our book was not intended to explain differential growth performance; it was aimed at explaining differential capacities for economic transformation. While the cases on which we concentrated—Ghana, Tanzania, Rwanda, and Ethiopia—were all fast growers, only Rwanda and Ethiopia, we argued, had a political-economic system conducive to the kinds of industrial policy required for sustainable development. 7 One implication is that growth in Englenbert and Portelance’s larger set of ‘developers’ is likely to be ephemeral, a prediction with which I suspect they might agree.

Finally, my reading of Englebert and Portelance’s charts is that there is a big FDI effect in a handful of cases, and much less of an effect in others. To me, this reinforces the belief that there are diverse explanations for growth among the ‘developers’, with the ‘signaling’ hypothesis more relevant to some countries than others. In any case, I look forward to their ongoing research which promises to shed additional light on these questions.

kelsall

Tim Kelsall is an independent academic and the former editor of African Affairs. His research focuses on political economy and political anthropology of Africa.

 


Goran Hyden

Englebert and Portelance’s piece is a refreshing look at some of the enigmatic issues in the study of African politics and development. I appreciate their attempt to move beyond conventional analyses. At the same time, as they acknowledge, the search for clarifications continues. How much of what we take for granted can and should be discarded and how much should be retained? What are the key challenges as we try to move beyond current paradigms and thinking? In this brief commentary I would like to raise four issues.

The first is how much we should trust our datasets? We tend to treat them as objective and rarely question how they are created. Not only is there a problem with much official statistics as Morten Jerven has pointed out. 8 The same applies to the democracy and governance indicators that members of the policy and academic communities have created. Especially problematic is Matt Andrew´s point about the tendency to raise scores based on announcement of policy reforms only to subsequently have them reduced as the record of implementation or enforcement becomes available. 9 It creates artificial up-and-down swings and constitutes a trap that needs to be carefully watched as further analysis of the growth-governance nexus is pursued.

The second point relates to whether democracy or good governance really explains development, or if it is the other way around. African politics fits into our theoretical and policy models like a square peg in a round hole. A quarter-century of scrutiny through governance and democracy models has only shown the limits of confining our research to regime rather than state issues. As the research from the African Power and Politics Programme at ODI suggests, the primary challenge to African development is not as much accountability as it is getting people to work together. How to pursue development tends to be a more urgent issue for policy-makers than how to produce democracy. Is it a coincidence that the two countries that have performed best – Ethiopia and Rwanda – are states with an indigenous monarchic legacy and lore? I suggest that, as we look beyond competing analyses based on economic and governance variables, we consider more thoroughly local African history and cultures.

Screenshot 2015-02-18 19.09.00

The third issue concerns the FDI lead. I don’t think that African leaders seriously adjust their governance to suit foreign investors. First of all, the latter tend to come regardless of quality of governance as long as there is a demand for a particular commodity. “One-stop-shops” are not really the true facilitators but a “golden handshake” often is. African notions of fairness are based on reciprocity, an issue that I raised in the 1992 book co-edited with Michael Bratton. 10 This lead was abandoned, however, in favor of analyzing governance issues in the context of democratic theory. The time has come to look at governance issues with greater attention to local notions of what is right and wrong. After all, African countries are in the process of building their own governance structures. Understanding this from the inside is still a challenge to political science research.

The fourth issue relates to how much economic growth figures tell us. What, if we look at the Human Development Index of the UN Development Program? A check on the average growth in human development between 2000 and 2013 for the countries included in Englebert’s and Portelance is presented in the table below:

hyden hdi chart

Table prepared by G. Hyden with HDI data from the UN Development Program.

This table confirms much of their findings but using data that are more closely related to how countries perform in terms of both economic and social development. It shows that the developers rank higher on the HDI and have achieved a greater average growth in human development than the laggards. The “star” performers – Ethiopia, Rwanda and Mozambique—are countries that have had a stable and authoritarian mode of governance; those among the developers with a higher ranking on the Freedom House scale – Botswana, Ghana and Cape Verde—have performed considerably less well. We do not have data for three countries among the laggards for the full period. However, as is evident from the table, they tend to be ranked lower than the developers. Their annual average growth is also generally low, with the exception of Burundi. In conclusion, the continuing search for explanations requires loosening our most comfortable theoretical blinders as they concern the state and modes of governance.

hyden

Goran Hyden is Distinguished Professor of Political Science at University of Florida. He studies the interface of politics and development, as well as democratization and governance.

 


Nicolas van de Walle

Pierre Englebert and Gailyn Portelance have given us much to chew on. Empirically based, nuanced, and skeptical of simplistic answers to the complex questions posed by African development, their study has a chance to reset the terms of the debate on the relationship between governance and development in contemporary Africa.

That the more successful countries in the region have better general governance than the less successful countries should not surprise us. I say not necessarily in a theoretical sense – since this seems controversial these days – but in the narrowly empirical sense that the relationship will be obvious to anyone who has recently paid visits to a ministry building, say in Ghana (one of their successful countries), and one in, say Guinea Bissau (one of their laggards).

More surprising, perhaps, is their second finding that the differences in governance are static and minor between the two categories of countries. Their explanation, that such minor differences might be enough to account for the much greater amount of FDI than the better-governed states get, is very suggestive.

This is a most welcome project and I look forward to seeing more of the research. In particular, it will be interesting to see what happens to these results when the authors move from bivariate to multi-variate analysis. The only potentially confounding factors they take into account are mineral rents and level of democracy, but there are plenty of others, from the level of GDP, to economic policy or terms of trade types of factors.

In addition, one wonders about endogeneity. 11 They assume, as do I, that governance explains growth more than the reverse, but one still wonders the extent to which the reverse causal sequence also matters. The extent to which there is growth, after all, probably shapes the government’s discount rate, and even its willingness to pay civil servants a living wage on a regular basis. In that case, perhaps a third factor “explains” the growth differential, which causes the higher FDI and improved governance.

Finally, I was struck that the level of FDI they find going to even the more successful states is relatively small. I would have liked to see it as a percentage of GDP, and disaggregated by country, since a nominal figure for the entire category of countries dissimulates more than it reveals. But, unless I misunderstand the figure, we are talking about lower levels than, say, remittances or foreign aid. I wonder if these FDI numbers could explain the differences in growth rates they have identified. I also suspect the differences in FDI are themselves endogenous to some other factor, which may be the key issue distinguishing these states.

vandewalle

Nicolas van de Walle is the John S. Knight Professor of Political Science at Cornell University. He studies democratization and the political economy of development, particularly in Africa.

 


Pierre Englebert & Gailyn Portelance

First of all, we thank Richard Joseph for giving our research such an outstanding platform and visibility. Second, we express our sincere gratitude to the four commenters: Jeffrey Herbst, Tim Kelsall, Goran Hyden, and Nicolas van de Walle. Frankly, it is hard to imagine a more impressive, knowledgeable, and thoughtful group of scholars. And, we should add, it is hard to imagine a kinder one.

Our research is ongoing and we are eager for feedback. We have not reached a stage of knowledge where we can firmly argue for particular positions. Thus, we find ourselves lucky to have such great feedback and intend to heed it as we move forward with this research.

There are broad lines of enquiry that the commenters suggest and which we will pursue: Jeffrey Herbst recommends that we move beyond FDI and look at signals from rulers to domestic audiences.Tim Kelsall rightly suggests that our hypothesis does not invalidate others, such as the emerging developmental statehood of some countries, and notes that the variance in the FDI effect deserves more attention.  Goran Hyden calls our attention to the quality of the data, and issues of endogeneity between governance and development. He invites us to think more critically about whether leaders really have FDI in mind when making governance decisions and to better accommodate local notions of governance.  Finally, Nicolas van de Walle, encourages us to move toward multivariate analysis, to pay more attention to endogeneity, and to question the surprisingly small size of our FDI figures with respect to their hypothetical growth effects.

We have our work cut out for us! Thank you.


  1. “Integrating Africa More Fully Into The Global Economy”, The International Monetary Fund, https://www.imf.org/external/np/sec/mds/1996/mds9610.htm. 
  2. See Akbar Noman et al., Good Growth and Governance in Africa: Rethinking Development Strategies (Oxford University Press, 2011). 
  3. For a review of Khan’s contentions as well as others who share this general line of approach, see R. Joseph, “Industrial Policies and Contemporary Africa: From Prebendal to Developmental Governance in Joseph E. Stiglitz et al., ed. The Industrial Policy Revolution II: Africa in the Twenty-First Century (Palgrave Macmillan, 2013). 
  4. Africa Power and Politics Programme, http://www.institutions-africa.org/. 
  5. “The Growth-Governance Paradox in Africa,” https://africaplus.wordpress.com/2015/01/06/the-growth-governance-paradox-in-africa/. 
  6. See R. Joseph, “Inclusive Growth and Developmental Governance: The Next African Frontiers,” in Célestin Monga and Justin Yifu Lin, eds., Oxford Handbook of Africa and Economics: Vol. 1 Concept and Contexts, online 2014, print 2015. http://oxfordindex.oup.com/view/10.1093/oxfordhb/9780199687114.013.29 
  7. See Tim Kelsall, Business, Politics, and the State in Africa (Zed Press, 2013); and also David Booth and Diana Cammack, Governance for Development in Africa: Solving Collective Action Problems (Zed Press, 2013). 
  8. Several of Jerven’s talks on this issue are available on YouTube. See https://www.youtube.com/watch?v=ypGikIFf8LM. 
  9. The Limits of Institutional Reform in Development: Changing Rules for Realistic Solutions (Cambridge University Press, 2013). 
  10. Governance and Politics in Africa (Lynne Rienner Publishers, 1992). 
  11. “Endogeneity” is given differing sets of meanings. It can mean “you got the causation wrong, that the model you wrote down and estimated does not properly capture the way causation works in the real world.” (Website: Cross Validated). Or, “in most cases where the charge of endogeneity is filed, people are not so much worried about omitted variables. Rather, what they are worried about are things like simultaneity, i.e., X causes Y but Y also causes X, and self selection.” (Jesper B. Sørensen) 
Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s