This chapter aims to identify the institutional areas that most constrain Benin’s economic development, relying on expert opinion as it appears in databases that provide international comparisons of institutional and related indicators, or as can be gathered locally through a dedicated survey. Section I of this chapter is devoted to systematic comparisons of Benin with other countries. Section II describes the survey carried out for the present study with a selected sample of decision makers in various areas and occupations in Benin. Section III synthesises the main lessons to be drawn from the preceding exercises and examines their degree of consistency with the conclusions of several growth diagnostic studies recently conducted on Benin.
I Benin’s Institutional Quality in an International Perspective
Three international databases will be used to compare Benin’s institutional quality with that of other countries. The first is the Quality of Government (QoG) database (Teorell et al., Reference Teorell, Dahlberg, Holmberg, Rothstein, Alvarado Pachon and Svensson2018). This is probably the most complete database available related to institutions. It comprises more than 2,000 indicators from more than 100 sources. Many fewer indicators are systematically available for low-income or lower middle-income countries – which are those that can be used as meaningful comparators for Benin. Nevertheless, there are still close to 200 indicators for such countries. In a previous use of that database (Bourguignon and Libois, Reference Bourguignon, Libois, Bourguignon and Wangwe2018), a clustering analysis has been employed to summarise this set of indicators into six synthetic indicators, based on the proximity of the inter-country profile of the original indicators they comprise.
The second database is the Worldwide Governance Indicators (WGI). This also relies on a wide collection of original databases. Instead of using a statistical method to aggregate all the individual indicators present in those databases, the aggregation is done a priori by broad governance areas and then the first principal component is extracted from country observations, which makes it possible to summarise the differences across countries in a single indicator of the quality of governance in that area (Kaufmann et al., Reference Kaufmann, Kraay and Mastruzzi2010). Six indicators covering different governance areas are derived in that way.
The third database is simply the Country Policy and Institutional Assessment (CPIA) indicators gathered annually by World Bank staff, re-aggregated in five broad clusters.Footnote 1
Benin’s institutional quality, as described by these three sets of synthesised indicators,Footnote 2 is compared against two groups of developing countries: neighbouring countries in Central and West Africa, and a group of countries that have performed better than Benin in terms of gross domestic product (GDP) per capita over the past decades despite being initially at a comparable level.Footnote 3 Ideally, the comparison of institutional quality should refer to that initial stage, to see whether countries that initially had better governance overall did better in a subsequent period. Such a historical comparison is possible (although somewhat problematic) with the WGI database, but not with the others.
A Benchmarking Benin against Neighbouring Countries
Benin shares direct borders with four countries, namely Togo, Burkina Faso, Niger, and Nigeria. We were, however, unable to include Niger and Togo in our comparison due to lack of data. We instead included other countries in the same geographical area: Cameroon, Côte d’Ivoire, and Ghana. Figures 3.1a–3.1f compare the institutional performance of Benin and its neighbouring countries using the three databases described earlier, and at different points of time for WGI and CPIA.
Figure 3.1a Governance synthetic indicators: Benin and its neighbours, 2015–2016. The reported figures represent simple averages of the scores for each country for the QoG indicators in 2015 and 2016
Figure 3.1b Governance synthetic indicators: Benin and its neighbours, 2018. The reported figures represent the score for each country for the WGI in 2018
Figure 3.1c Governance synthetic indicators: Benin and its neighbours, 2005. The reported figures represent the score for each country for the WGI indicators in 2005
Figure 3.1d Governance synthetic indicators: Benin and its neighbours, 1996. The reported figures represent the score for each country for the WGI in 1996
Figure 3.1e Governance synthetic indicators: Benin and its neighbours, 2016–2017. The reported figures represent simple averages of the scores for each country for the CPIA indicators in 2016–2017
Figure 3.1f Governance synthetic indicators: Benin and its neighbours, 2005. The reported figures represent simple averages of the scores for each country for the CPIA indicators in 2005–2006
Even though the synthesised indicators constructed based on the three databases often have similar names (Figures 3.1a–3.1f), they do not necessarily cover the same concepts. This would be the case for the indicator entitled ‘control of corruption’. In the QoG database, this indicator is combined with the equal implementation of the law, which is part of the ‘rule of the law’ in the WGI and the CPIA. Likewise, human rights in the QoG refer to personal liberties, but also to the provision of public services to individuals, including education, health care, or social assistance, something that is behind the ‘people’ label in the CPIA database. Other indicators are conceptually closer to each other even though they have not been given the same name. This is the case for ‘business environment’ (CPIA), ‘regulatory quality’ (WGI), or ‘private-sector competitiveness’ (QoG). This is also the case for ‘government effectiveness’ (WGI), ‘administrative capacity’ (QoG), and ‘public management’ (CPIA). The same can be said of ‘democratisation’ (QoG) and ‘voice and accountability’ (WGI).
With the precaution required by this heterogeneity of indicators attached to different databases, we now examine the kind of differences in the quality of institutions they suggest exist between Benin and neighbouring countries.
The convergence across databases is stronger than may be apparent at first sight. Three synthetic indicators appear at least twice as relative weaknesses in the 2015–2016 data: business environment (QoG and CPIA), government effectiveness (QoG, CPIA), and control of corruption (QoG and WGI). On the side of the relative strengths of Benin, voice and accountability and human rights are unanimously better than in the comparator countries, the same being true of political stability in WGI or the absence of conflict and violence in the QoG.
The lack of full convergence in areas that seem to be well defined across the three databases may seem surprising. As mentioned earlier, however, the concepts behind the corresponding synthetic indicators may differ. For instance, new policies to control corruption may be praised in the CPIA corruption indicator, whereas other databases focus on the fact that the level of corruption remains unchanged. Likewise, some indicators may stress structural obstacles in ‘doing business’, like insufficient infrastructure, whereas others will put more emphasis on the government’s attitude towards business. The cost of relying on synthetic indicators is precisely that it is not possible to get into this kind of detail, this being the reason why an institutional diagnostic must necessarily go beyond this kind of aggregate analysis.
The comparability over time of the WGI is somewhat uncertain because the number of databases used to build them has substantially expanded over the last two decades. Yet, the relative position of countries along the various indicators should not be too greatly affected by this problem. From that point of view, no noticeable change in the ranking of Benin took place over these two decades, except for ‘regulatory quality’ (i.e. business environment), where Benin tends to progressively lag behind Ghana and Burkina Faso over time.
The comparability over time of the CPIA governance quality indicators is probably better than for WGI because they are supposedly based on a homogeneous methodology. There, the most noticeable change is again the worsening of the business environment both in absolute terms and relatively to neighbouring countries.
Overall, the appraisal of the quality of institutions in Benin through aggregate indicators and the comparison with neighbouring countries points to three weaknesses: the control of corruption, the business environment, and public management. Benin does not exhibit the worst performance in these areas at any point in time, as Cameroon and Nigeria most often lie behind it. However, it is generally the case that Benin does not do as well as Ghana, which dominates all the other countries in 2015–2016 according the WGI, or as well as Burkina Faso. Over time, moreover, it would seem that regress rather than progress is observed in the business environment.
B Benchmarking Benin against Better-Performing Developing Countries
We now compare Benin with five developing countries whose level of economic development was similar to Benin in the early 1990s, but that have had higher per capita GDP growth rates over the past twenty-five years and have now become substantially richer than Benin. These are Bangladesh, Cambodia, Lao, Vietnam, and Tanzania. Figures 3.2a–3.2f present the comparison using the same three sets of indicators in the same periods as in Figures 3.1a–3.1f. Of course, the profile of Benin in all radar charts is the same. What matters now is how Benin compares to those countries that were able to grow faster, both today and in the past at a time when all of the countries were at a comparable level of GDP per capita.
Figure 3.2a Governance synthetic indicators: Benin vs better-performing countries, 2015–2016. The reported figures represent simple averages of the scores for each country for the QoG indicators in 2015 and 2016
Figure 3.2b Governance synthetic indicators: Benin vs better-performing countries, 2018. The reported figures represent the score for each country for the WGI in 2018
Figure 3.2c Governance synthetic indicators: Benin vs better-performing countries, 2005. The reported figures represent the score for each country for the WGI in 2005
Figure 3.2d Governance synthetic indicators: Benin vs better-performing countries, 1996. The reported figures represent the score for each country for the WGI in 1996
Figure 3.2e Governance synthetic indicators: Benin vs better-performing countries, 2016–2017. The reported figures represent simple averages of the scores for each country for the CPIA indicators in 2016–2017
Figure 3.2f Governance synthetic indicators: Benin vs better-performing countries, 2005. The reported figures represent simple averages of the scores for each country for the CPIA indicators in 2005–2006.
Looking first at the three radar charts in Figures 3.2a–2c, the common feature is that, in comparison to these better-performing countries, Benin does not do well in public management and private-sector competitiveness. Except for this, Benin turns out to be quite comparable to the other countries when considering the WGI. It even performs relatively well on corruption. The latter advantage is still more pronounced with respect to the CPIA indicators. Overall, it would thus seem that, paradoxically, better-performing countries do systematically worse than Benin on that account.
Things are a bit different with the QoG database. There, Benin’s control of corruption appears to be among the worst. In the QoG database neighbouring countries’ control of corruption are also worse than they appear according to the WGI and CPIA indicators, confirming the different understandings of the concept of ‘corruption control’ in the various databases. The same is true of business environment in the QoG database, where Benin performs rather worse than comparator countries, whereas it ranks at or close to the median in the other databases.
Another somewhat surprising result is the underperformance of Benin according to the ‘people’ indicator in the CPIA database. The point here is that, under the general heading of social inclusion and equity, that indicator puts a great weight on education. The interpretation to be given to the gap observed for Benin is therefore that fewer or lower-quality public efforts are devoted to human capital accumulation. Better-performing countries have invested more than Benin in that institutional aspect of development. This was not systematically the case when the comparison was with neighbouring countries.
Looking now at the radar charts for earlier periods, Benin’s human capital gap is fully confirmed on the CPIA for 2005. It is also noticeable that over the last decade or so Benin has lost the comparative advantage it initially enjoyed in its business environment relative to better-performing countries. This finding raises the issue of whether initial institutional advantages, as measured by the kind of synthetic indicator used here, are responsible for the faster development of these countries, or whether it is their development that created such advantages.
One reaches the same conclusion when looking at the three WGI charts. Clearly, Benin was, roughly speaking, doing better than other countries in 1996. In particular, its radar profile dominated that of Bangladesh, Cambodia, and Tanzania in all areas. If it had not been for corruption, it would also have dominated Lao and Vietnam. Paradoxically, however, all of these countries grew much faster than Benin in the subsequent twenty years.
C Conclusion on International Comparison
What lessons may be drawn from this review of available international governance and institutional synthetic indicators? The first must be the lack of convergence of indicators from different databases that are nevertheless supposed to cover comparable areas. This discrepancy can only be explained by heterogeneous conceptual definitions, but it also casts some doubt on the true meaning of any single synthetic indicator of the type so frequently used in the cross-country development literature. Being analytically more rigorous would require using much more precise indicators, but this would increase the number of indicators to be used and would add to the inconclusiveness of the analysis.
Second, concerning Benin, the three potential sources of institutional weakness revealed by the analysis are the control of corruption, a business environment that is possibly less favourable than that in other countries, and efficiency issues in public management (although cross-country differences were rather small on that latter account). Another important weakness seems to lie in the public investment in ‘people’, most likely due to an underperforming educational system.
These are extremely general conclusions and, therefore, of limited use for policy makers. Remedying this would require getting into more detail to try to identify what exactly is making the business environment unfavourable or public management ineffective. As mentioned earlier, however, multiplying the number of dimensions of this type of international comparison would quickly render any results impossible to interpret. Other approaches must be developed to make use of the general indications delivered by the preceding analysis. Such approaches are pursued in the rest of this chapter and the accompanying chapters of the diagnostic.
II A Survey of Experts’ Opinions on Benin’s Institutional Quality
The goal of this section is to derive further insights about the quality of institutions in Benin on the basis of opinions obtained from decision makers in the private and public sectors, as well as from civil society, who are directly exposed to these institutions. An opinion survey has been conducted with a sample of such people. The questionnaire was adapted from the one applied in a similar study in Tanzania to fit the reality of Benin. In the following paragraphs we first describe in some detail the methodology of this survey, before analysing its results and then underlining the lessons to be drawn in terms of institutional strengths and weaknesses of development in Benin.
A Methodology
The survey of experts’ opinions about Benin’s institutional performance was developed in collaboration with Analysis for Economic Decisions, a Belgian consultancy, and a local team led by the director of Benin’s National Institute of Statistics and Economic Analysis, under the close supervision of the authors of the present chapter. The methodology included three main steps: questionnaire, sampling, and survey implementation.
1 The Format of the Questionnaire
The questionnaire was adapted from a similar survey carried out in Tanzania – see Bourguignon and Libois (Reference Bourguignon, Libois, Bourguignon and Wangwe2018), after translation into French and modifications required by the Benin context. Questions that were irrelevant to Benin were excluded, and new questions were added based on insights from the first chapters of this volume and a workshop with key decision makers that took place in August 2017 in Cotonou. A number of questions were also reformatted so as to facilitate communication during interviews. Finally, the questionnaire was coded into Survey CTO, allowing it to be implemented on tablets.
The format of the questionnaire is somewhat original. It was initially conceived to cover most economic, political, and social institutional issues. As it was too long for a single respondent, and because all respondents would not be knowledgeable in all areas, a flexible format was adopted, where respondents would choose the areas they would focus upon. By doing so, however, they would reveal at the same time their opinion about the strength of the institutional constraints on development in the various areas they could choose from.
Practically, the questionnaire consists of ten subsets of questions, each one corresponding to a broad institutional area: political institutions, law and order, ease of doing business, public administration, and so on. The list of areas appears in the working paper version (Houssa and Bourguignon, Reference Houssa, Bourguignon, Bourguignon, Houssa, Platteau and Reding2019). First, respondents were asked to indicate which three of the ten institutional areas they saw as including the most constraining factors for Benin’s economic development. Second, they were asked to give a relative weight to these three critical institutional domains, where a high value assigned to an area indicated that it is more detrimental to Benin’s economic development. Then, respondents had to answer those questions in the questionnaire that came under each of their three critical area headings, plus a fourth area chosen randomly in order to make sure that all questions in the questionnaire would be answered a minimum number of times.
2 Sampling
The survey team developed a sampling strategy that relied on a demand-side/supply-side approach to analysing institutions. First, public or private entities – firms, public administrations, agencies, political parties, trade unions, and so on – were identified, some of them being involved in setting or managing institutions, whereas others were simple users of those institutions in their customary activities. Then, respondents were selected within these entities, preferably among senior managers or deputies.
The sample comprises 396 respondents across five key groups of entities/experts, summarised in Table 3.1: public administration, judiciary, executive and legislative bodies, donors, civil society, and the private sector. Each group was further divided into subgroups with possibly a different relationship to similar institutions. For instance, the private-sector group includes three subgroups: formal firms, informal firms, and financial institutions; and public administration includes sectors like education, health, or utilities.
Table 3.1 Overview of the sample
| Category | No. | % | Category | No. | % |
|---|---|---|---|---|---|
| Public sector: total | 131 | 33 | Members of parliament | 23 | 6 |
| Public administration | 92 | 23 | Members of other constitutional bodies (supreme court, auditor general, …) | 27 | 7 |
| Agriculture, commerce, industry | 19 | 5 | Trade unionists | 3 | 1 |
| Energy, water, mining | 6 | 2 | Donors | 9 | 2 |
| Economy, finance, development | 19 | 5 | Civil society | 24 | 6 |
| Education | 11 | 3 | Academics | 10 | 3 |
| Health | 9 | 2 | Think-tanks and charitable organisations | 4 | 1 |
| Infrastructure, transport, communications | 14 | 4 | Media | 10 | 3 |
| Sport, culture, tourism | 8 | 2 | Private sector | 170 | 43 |
| Foreign relations | 6 | 2 | Formal private firms | 82 | 21 |
| Law and order | 15 | 4 | Large firms and their associations | 49 | 12 |
| Judiciary | 7 | 2 | Medium firms | 10 | 3 |
| Military | 4 | 1 | Small firms | 6 | 2 |
| Police | 4 | 1 | Micro firms | 17 | 4 |
| Other administrations | 24 | 6 | Finance | 12 | 3 |
| Executive | 7 | 2 | Banks and their associations | 6 | 2 |
| Retired ministers | 11 | 3 | Micro-finance institutions | 6 | 2 |
| Local administrations (départements) | 6 | 2 | Informal firms | 76 | 19 |
| Political institutions | 62 | 16 | Total | 396 | 100 |
| Local politicians (communes) | 9 | 2 |
Two methods were used to select entities in each subgroup: an arbitrary selection and a random sampling approach. Arbitrary selection was used to select entities in official sectors; that is, public administration, political institutions, civil society, or donors. Geographical diversity (départements and communes) was also taken into account as much as possible.Footnote 4 A random sampling approach was implemented to select entities within the private-sector subgroups – except for the financial sector, where specific executives were arbitrarily selected, for the same reason as in the public sector; that is, the reduced number of entities to be considered.
Two specific strategies were used to randomly select formal and informal private firms. On the one hand, a database (Déclarations Statistiques et Fiscales, which includes 5,361 firms) was used to randomly select around eighty formal firms according to firm size, after stratifying the universe by size, but irrespectively of economic sector of activity. On the other hand, seventy-five informal firms were randomly selected after stratifying by sector of activity and geographical area. The random selection was made by enumerators who had been assigned a location and a field of activity.
3 Survey Implementation
The survey was implemented between December 2017 and early February 2018. Map 3.1 displays the locations of respondents’ entities.
Map 3.1 Geographical origins of respondents’ entities
The map shows that respondents are spread across the country. However, it turns out that the Atacora département is not represented in the sample. Also, there is an over-concentration of respondents in the southern (Ouémé, Atlantique, and Littoral) and north-eastern (Borgou) parts of Benin – the départements where most of the arbitrarily selected entities are located. In particular, the city of Cotonou in the département of Littoral is home to many of the firms and governmental entities. It must be kept in mind, however, that the objective of the survey is to poll not the Beninese population, but people with some knowledge and experience of the way various types of institutions function in Benin.
The stratification by type of occupation and geographical areas reflecting this choice does not necessarily fit the geographical distribution of the population. The sample of respondents is not representative of the Beninese population. In the sample, 84 per cent have a university degree and 27 per cent have studied abroad. They are in their mid-40s on average, and most of them have a family. Perhaps because of the education bias, Christians are over-represented in comparison with the whole population. If there is no strong bias in terms of ethnicity, there is in terms of gender: the sample is strongly dominated by males (82 per cent). This feature reflects the gender distribution among senior managers in Benin. As a matter of fact, the only sector where a gender balance holds is among respondents operating in the informal sector (see more detail in Houssa and Bourguignon, Reference Houssa, Bourguignon, Bourguignon, Houssa, Platteau and Reding2019).
B Empirical Results
This section summarises the information derived from the expert opinion survey. This is done into two steps. First, we analyse responses to the question that three broad institutional areas, among the ten areas listed in Table 3.2, are the most constraining for the development of Benin overall. Second, we gain a deeper understanding of the reasons behind those choices by analysing the responses to the specific questions that come under each broad area heading.
Table 3.2 Broad institutional areas by perceived weaknesses
| Total | First choice | Second choice | Third choice | ||||||
|---|---|---|---|---|---|---|---|---|---|
| No. | % | Rank | No. | % | No. | % | No. | % | |
| A. Political institutions | 201 | 16.9 | 2 | 170 | 42.9 | 17 | 4.3 | 14 | 3.5 |
| B. Law and order, justice, and security | 126 | 10.6 | 4 | 40 | 10.1 | 63 | 15.9 | 23 | 5.8 |
| C. Public administration | 230 | 19.4 | 1 | 76 | 19.2 | 110 | 27.8 | 44 | 11.1 |
| D. Ease of doing business | 133 | 11.2 | 3 | 32 | 8.1 | 53 | 13.4 | 48 | 12.1 |
| E. Dealing with land rights | 127 | 10.7 | 4 | 29 | 7.3 | 48 | 12.1 | 50 | 12.6 |
| F. Long-term and strategic planning | 85 | 7.2 | 7 | 12 | 3 | 38 | 9.6 | 35 | 8.8 |
| G. Market regulation | 83 | 7 | 7 | 15 | 3.8 | 39 | 9.9 | 29 | 7.3 |
| H. Security of transactions and contracts | 21 | 1.8 | 10 | 2 | 0.5 | 8 | 2 | 11 | 2.8 |
| I. Relations with the rest of the world | 63 | 5.3 | 9 | 6 | 1.5 | 9 | 2.3 | 48 | 12.1 |
| J. Social cohesion, protection, and solidarity | 119 | 10 | 6 | 14 | 3.5 | 11 | 2.8 | 94 | 23.7 |
| Total | 1188 | 100 | 396 | 100 | 396 | 100 | 396 | 100 | |
1 Perceived Institutional Constraints by Broad Areas of Functioning of Institutions
The ‘total’ row in Table 3.2 reports the number of times each broad institutional area appeared among the three most critical areas for Benin’s development mentioned by respondents. Two areas strongly dominate the others: the functioning of public administration, followed by the functioning of political institutions. Together, they account for one-third of all opinions. Some way after them comes a group of four other areas that each account for about 10–11 per cent of the total choices: law and order, justice, and security; ease ofdoing business; land rights; and social cohesion, protection, and solidarity. The four remaining areas – market regulation, long-term and strategic planning, security of transactions and contracts, and relations with the rest of the world – seem to be less critical. This may be because they are seen as corresponding to more technical aspects of institutions, and therefore were probably more distant from the preoccupations of respondents.
Equally interesting is the order of appearance of each institutional area in the choice of three areas by the respondents. It can be seen in Table 3.2 that political institutions was mentioned by 43 per cent of the respondents as their first choice, followed by public administration, which also dominates the second choice. The third choice is dominated by social cohesion, protection, and solidarity. This result must be interpreted negatively, though. Indeed, that area (social cohesion, protection, and solidarity) appears to be of less importance in comparison with areas (e.g. areas B, D, and E) that have more or less the same total number of mentions, but that were mentioned more frequently as the first and second choices.
The preceding ranking is changed only marginally when the weights respondents associated with the institutional areas they selected are accounted for. Political institutions and public administration remain strongly dominant. As a matter of fact, the average weight given to the institutional areas by those respondents who mentioned them as an obstacle to development is quite uniform, except, interestingly, for political institutions, which again dominates the others. The same results were obtained when respondents were asked to reveal their willingness to pay for improving those institutions they found most critical for development (see more detail in Houssa and Bourguignon, Reference Houssa, Bourguignon, Bourguignon, Houssa, Platteau and Reding2019).
Respondents are expected to have heterogeneous views about institutional weaknesses. In order to gain insights into this issue, a number of mechanical regressions were run where a dummy variable defined by whether a broad institutional area was seen as critical or not was regressed on some characteristics of respondents, namely being a Beninese national; being a woman; managing a large, medium, or small firm; and being employed in a financial institution.
These regressions are shown in Table 3.A.1 in the Appendix to this chapter. Among the noteworthy results is the fact that large and formal firm operators tend to have less distrust than other respondents with respect to political institutions – perhaps because they know better how to deal with them. The other side of the coin is that, more than others, they find that the business environment, including market regulation or the security of contracts, is an impediment to development. This attitude is still more prevalent among respondents working in financial institutions. More surprisingly, women also share this view; that is, they place less emphasis on political institutions and more emphasis on business, possibly because they tend to be over-represented among small and micro entrepreneurs. As far as nationality is concerned, it is not clear that the distinction is meaningful given the tiny minority of foreigners in the sample. Not surprisingly, foreigners overvalue the business environment whereas nationals give more importance to land rights.
2 Interactions between Formal and Informal Institutions
In a society where tradition very much matters, it was considered interesting to ask respondents about whether traditional institutions could be a good substitute for imperfectly working formal institutions, particularly those institutions dealing with business relationships, for instance security of contracts (Dhillon and Rigolini, Reference Dhillon and Rigolini2011). In this regard, respondents had to choose out of five informal institutions those they considered to be a good substitute for imperfectly working formal institutions: religious leaders, traditional authorities, networks, other personal relations, and cultural masking traditions established during the pre-colonial period and backed by spiritual forces.Footnote 5
The responses suggest that the dominant informal response to institutional weaknesses are not the traditions inherited from pre-colonial times, but essentially private networks and, to a lesser extent, traditional and religious leaders. This result was not unexpected given the rather high average educational level of the sample. Yet, the fact that a significant proportion of respondents mentioned traditional and religious leaders is evidence that formal institutions regulating interpersonal economic relationships are not fully established in Beninese society, possibly because of the survival of traditional means of solving this kind of problem.Footnote 6
3 In-Depth Perceptions of the Quality of Institutions in Benin
We now go one step further by exploiting the detailed questions asked of the respondents in connection with the three broad areas they chose, and a randomly selected one. The full list of questions may be found online.Footnote 7 For the sake of simplicity, however, we shall not deal with these questions directly. We shall rather list the main lessons that can be learned from the answers. Before doing so, however, we must address some methodological issues in the identification of weaknesses and strengths revealed by the answers to the questionnaire.
The response to all questions was coded on a Likert scale ranging from 0 to 4: 0 defined as ‘not at all’, 1 as ‘little/low importance’, 2 as ‘neutral’, 3 as ‘a lot’, and 4 as ‘extremely’. In addition, respondents were allowed to reply with ‘I do not know’ when they could not provide relevant answers to a question. Note that some questions were asked in a negative way – for example, ‘To what extent does corruption constrain business?’ – whereas others were asked in a positive way – for example, ‘How well do you think local communities understand aspects of the land law that concern them?’ To make responses comparable across the questions, the answers were re-coded to the negative, so that all low response values can be interpreted as institutional weaknesses, and high values as strengths.
The full questionnaire is very rich, as it includes more than 400 questions – even though the typical respondent had to answer roughly half of them; that is, those in the areas he/she chose. To synthesise the answers, a number of methodological choices have to be made.
Weaknesses and strengths are defined by average Likert scores below 1.5 for the former and above 2.5 for the latter. These cut-offs were defined on the basis of the distribution of average scores across all questions shown in Figure 3.3a, which exhibits discontinuities at these values. Note, however, that relying on average scores raises the issue of how to interpret responses with neutral opinions; that is, those with scores equal to 2.Footnote 8 Is it a truly ‘neutral’ response or a quick way to get rid of a question one cannot really answer? To take this ambiguity into account, questions were ranked in accordance with both their average score and that score after eliminating the 2-scores. However, the difference between the two rankings was marginal. The same was found when considering only the proportion of scores strictly below 2 for weaknesses and strictly above 2 for strengths.
Figure 3.3a Average scores: Distribution of average scores
A first consistency check of this methodology of handling the answers to the various questions in the questionnaire consists of checking whether the scores of the questions under the heading of the ten broad institutional areas fit the average ranking done by respondents in the first part of the interview. This is done in Figure 3.3b.
Figure 3.3b Average scores: Frequency of questions by score levels
The first block on the left-hand side of Figure 3.3b simply shows the relative frequency of questions across broad institutional areas in the questionnaire. For instance, about 30 per cent of all questions fall under the heading of ‘political institutions’. However, it must be kept in mind that some questions appear under various headings. For instance, a question on the corruption of tax collectors would appear under both public management and ease of doing business.
The two other blocks of the right-hand panel of Figure 3.3b are more interesting. They show the same frequencies, but now restricting the universe to questions whose average score is below 1.5 (i.e. weaknesses) in the middle block and above 2.5 (strengths) in the right-hand block. What is interesting here is that the relative weakness of the broad areas is now slightly modified in comparison with the direct ranking operated by the population of respondents.
It is still the case that public management is considered to be the weakest area, since the frequency of questions with an average score below 1.5 is higher than the frequency of all questions in that area – and of course the frequency of questions with scores above 2.5 is lower. Yet, the second weakest area now appears to be the ease of doing business, for which the same pattern holds. By contrast, political institutions, which were considered practically as bad as public management in the direct ranking (i.e. when no detail was given to the respondents about what precise issue this area was covering), now would be more on the positive side: this area shows relatively more questions with high scores and fewer with low scores. This means that asking people about institutional weaknesses and strengths without first making them aware of what each area actually covers may be misleading. In the case of Benin, there are of course severe weaknesses in the way the political institutions work, but respondents also point to very positive aspects, so that, overall, their opinion is certainly not as negative as when asked whether ‘political institutions work well or badly’ without further detail. This is not true, however, of public management and doing business, which are still considered to be major institutional weaknesses.
Three other areas show some reversal of opinion when detailed aspects of the institutional area are given to respondents. The first is social cohesion, protection, and solidarity, where questions with high scores strongly dominate, and long-term planning, where the opposite is the case. In the former, the problem may come from the fact that the title of the area comprises different concepts and it is not clear which one dominated in the mind of respondents when first confronted with it. It is possible, for instance, that they may have put more emphasis on social protection, which they consider to be a weakness, and then realised when faced with the detailed questions that this area was also about traditional solidarity among people, which they considered to be a strength. For ‘long-term and strategic planning’, it is also probable that the understanding of that label was modified when respondents realised what it referred to. The third area that appears weaker than it was initially is land rights, where the frequency of low-score questions is twice that of questions about land rights overall.
In summary, the more detailed questionnaire showed some change in the ranking of broad institutional areas by relative weakness or strength. The main weaknesses revealed by the questionnaire are public management and ease of doing business, but, of course, it is now necessary to obtain deeper insights by focusing on individual questions and examining in more detail those with the lowest and highest scores. This is done in the next sections, which look successively at the revealed institutional weaknesses and strengths.
4 Perceived Weaknesses of Institutions
Instead of analysing one by one all the questions in the questionnaire that received an average score of less than 1.5, we list in what follows the main lessons that can be learned from them. Because the questions address a large range of issues, this approach allows for a more detailed diagnostic of institutional weaknesses than simply ranking broad institutional areas, as was done earlier. Tables 3.3a and 3.3b report the results for institutional weaknesses and strengths, respectively.
Table 3.3a Selected examples of detailed institutional performance: weaknessesa
| Questions on weaknesses | Average score |
|---|---|
| To what extent is corruption an obstacle to business development? | 0.73 |
| Do you think the Haute Cour de Justice is able to impose the respect of the constitution? | 0.76 |
| What is the degree of corruption linking the media and politicians? | 0.82 |
| What is the degree of political corruption (vote buying, illegal campaign funding, bribes)? | 0.83 |
| Does the government discuss the budget seriously with the civil society? | 0.87 |
| To what extent do land transactions involve corruption in local communities? | 0.9 |
| How much would you say the press and the media are independent from political influence? | 0.96 |
| Do you trust the Haute Cour de Justice to impose the legal rules of the game on the main political and economic actors? | 0.98 |
| To what extent are public procurement procedures fair and transparent? | 0.98 |
| In your view, is poverty reduction a priority for political parties? | 0.99 |
| What is the degree of corruption in the relationship between public administrations and Beninese companies? | 1 |
| To what extent is the reliability of economic aggregates like GDP growth, the current account balance, or inflation discussed in parliament, in the media, and in the civil society? | 1.15 |
| How seriously are public accounts audited? | 1.18 |
| To what extent are political dissensions obstacles to the implementation of public policies and reforms? | 1.22 |
a Recall that scores are redefined depending on the question, so that a low score denotes an institutional weakness. For instance, if the answer to the first question ‘To what extent is corruption an obstacle to business development?’ is ‘very much so’, and thus a Likert score of 4 is applied, it is redefined as 0 in agreement with the fact that this denotes a major obstacle to development. The reported score is the average 0/4 Likert scores after eliminating scores equal to 2.
Table 3.3b Selected examples of detailed institutional performance: strengthsa
| Question on strengths | Average score |
|---|---|
| Does the state discriminate among citizens in regard to accessing administrative services, justice, security, public school, health-care centres, etc.? | 3.6 |
| To what extent is the army or the police involved in politics? | 3.29 |
| In your opinion, is wage discrimination with respect to religion or ethnicity frequent in the private sector? | 3.17 |
| How free do you feel people are to form associations of a religious, ethnic, professional, or political nature? | 3.11 |
| Did the one-stop shop policy recently implemented in public administration improve doing business? | 2.91 |
| In view of religious, traditional, or ethical norms, how free is the Benin state about policies and reforms in education, health, social services, and economic policy? | 2.9 |
| How strong is the national sentiment in Benin? | 2.87 |
| Are traditional solidarity links effective in supporting people in need in rural areas? | 2.8 |
| How repressive do you feel the Benin state is? | 2.74 |
| Do you think that present reforms in the anti-corruption policy will lift constraints on development? | 2.7 |
a Recall that scores are redefined depending on the question, so that a low score denotes an institutional weakness. For instance, if the answer to the first question ‘Does the state discriminate among citizens in regard to accessing administrative services, justice, security, public school, health-care centres, etc.?’ is ‘very much so’, and thus a Likert score of 4 is applied, it is redefined as 0 in agreement with the fact that this denotes a major obstacle to development. The reported score is the average 0/4 Likert scores after eliminating scores equal to 2.
Without doubt, corruption is the theme that appears most frequently among the questions that obtained the lowest average scores among respondents. It affects practically all aspects of political and economic life in Benin: the political system, the relationship between business and the public administration (rigged procurement) or the judiciary system, the electoral system (vote buying), land rights, or complicity between politicians and the media. Corruption is seen as responsible for several key dysfunctions in the political, judicial, and economic spheres.
Another problem that is frequently mentioned, which also relates to corruption, or more exactly the difficulty of controlling it, is that the official rules of the political game, namely the constitutional rules, may be violated without the entities supposed to punish such behaviour taking action. The Constitutional Court, the Supreme Court, and the Haute Cour de Justice, whose jurisdiction is the illegal behaviour of the executive, are generally found to be permissive or passive, the same being true of the parliament. It is quite possible, however, that this opinion among the respondents was strongly influenced by the debate at the time the survey was taken about several decisions by the incoming president, which some felt were in contradiction with the constitution (see for instance Hessoun, Reference Hessoun2017).
The lack of transparency of state decisions and state action is another theme that attracts low scores. For instance, the criticism is made that no public discussion takes place about the execution of the budget or National Accounts, that the financial results of public and semi-public companies are not made public and not debated, that most decisions by the executive are taken in an opaque way, and that few evaluations are made of policies.
A theme that is of importance is the understanding that citizens have of the law and the rules of the game. There was a single question addressing this issue in the questionnaire and it referred to land law. The general opinion in this respect was that local communities have a poor understanding of the law and cannot use it to protect themselves against illegal practices that would take the control of some land away from them.
Concerning state-owned companies, their efficiency and management were severely criticised by respondents. This was especially the case for the company responsible for the production and distribution of electricity.
Two additional points that are apparent in the responses to the questionnaire are worth stressing. The first is the low average score for the question about whether poverty reduction could be considered as the main objective of policy making in Benin. The second is the view that dissension does exist within the executive itself. Here again, however, it may be the case that the low score for that question was influenced by some specific event that took place during the time of the survey or a little before – despite the fact that respondents were explicitly asked to base their answers on the way they saw politics, economics, or the working of the administration over the ten years preceding the survey, rather than basing it on current events and debates.
Reported development bottlenecks also include the dominant informal sector, Benin’s dependency on Nigeria, labour market nominal wage rigidity, and the frequent strikes in the public sector. All these constraints generate high costs for businesses and undermine competitiveness. However, it is not clear that they refer to institutional weaknesses strictly speaking.
5 Perceived Strengths of Institutions
Strengths are supposed to be revealed by questions with a score above 2.5; that is, a majority of respondents having selected the top value on the Likert scale. The main points that arise from reviewing these questions are the following.
High levels of respondent satisfaction mostly centre on five domains, which are not always fully consistent with perceived institutional weaknesses. These are (1) civil liberties; (2) a sense that the state enjoys some autonomy in policy making; (3) some trust in recent reforms; (4) a feeling of improvement in the ease of doing business; and (5) national pride.
On civil liberties, respondents expressed satisfaction with respect to the freedom given to people to form associations in practically all areas, from religion to politics. Equally important was the feeling of limited religious, ethnic, and political discrimination in recruitment and wage practices in the private sector. The lack of discrimination in access to public services – schools, health facilities, justice, or security – was also highly valued. Consistent with these civil liberties, the lack of state repression was also stressed by respondents.
The autonomy that respondents feel the Beninese state enjoys with respect to social, traditional, ethnic, and religious norms, or with respect to the army and the police, is certainly an advantage over some other countries. Yet, this feeling may not be fully consistent with the importance of corruption so strongly emphasised among key institutional weaknesses. In other words, autonomy does exist with respect to some norms and some specific actors, but it is probably more limited when dealing with big business or some other vested interests.
The prevalence of corruption among the perceived institutional weaknesses of Benin was such that it is somewhat surprising that respondents tend to trust announced anti-corruption reforms. Or is it precisely because corruption has reached such a critical level that experts tend to agree on the need to fight it effectively? The confidence expressed in the positive impact of aid, or at least on the absence of the crowding-out effect of aid on domestic savings, is also unexpected at a time when aid effectiveness is increasingly open to doubt. Yet, one may understand why such a point of view prevails in an economy where aid represents between 6 and 8 per cent of gross national income.
Recent reforms seem to have improved the way business feels about the business environment, even though it was seen earlier that there were still many causes of dissatisfaction. The one-stop window for formalities and the shortening of registration delays were sources of satisfaction for business-oriented people among the respondents. The low probability of violent events or worker strikes in the private sector was also felt to be a positive aspect of the business environment.
Finally, the feeling of belonging to a national community may not be easily related to the broad institutional areas that have been discussed in this chapter. That it appears with a strong score in the questionnaires despite the ethnic diversity of the country is a positive sign: the probably of conflict and violence is therefore reduced, which should be favourable to business and long-term public planning.
6 Perceived Opinion on Recent Reforms
Several reforms were recently initiated by the Talon administration, some of them with the ambition of improving the institutional framework of Benin’s development. Respondents were initially asked to answer questions based on their knowledge and experience over the preceding ten years, which is mostly before the Talon administration came to power. This was done in order to have a picture of expert opinion on Beninese institutions that would not be biased by the debate about the most recent reforms. Because of this, it seemed interesting to ask the experts briefly about these reforms, to check whether their views would differ.
Four types of reform were launched by the new administration. The first consisted of moving activities initially under the responsibility of civil services to agencies formally outside the public sector. Their mission is the same, but they escape some of the constraints of operating in the public sector, thus making them potentially more effective. For instance, agencies were created to manage the construction of schools and health centres, and others were created to replace the public company Société Nationale pour la Promotion Agricole (SONAPRA), which was responsible for agricultural promotion, rural development, and price stabilisation; another agency was created to manage water projects, and so on. These are potentially major reforms. It is of course too early to evaluate the reforms’ impact, but it is interesting to note that survey respondents were essentially either neutral or ambivalent with respect to them. Indeed, the average score for the questions about these reforms was very close to 2, and roughly 40 per cent of respondents either reported 2 or did not answer the question.Footnote 9
On a more positive side is the recent law that strengthens the land reform undertaken over recent years, and in particular the land titling operation launched in 2013 with the help of the US Millennium Challenge Account programme. One problem with the ongoing reform, however, is that a land title does not provide a definitive right until after five years, and it may be contested during this entire period. Indeed, several such contestations have taken place, and financial institutions that use land for collateral have experienced losses. As a result, they have become reluctant to accept land with temporary rights as collateral. To address this issue, the Talon administration passed a new law in late 2016 that gave landowners definitive rights. Survey respondents supported this reform, more strongly it should be said than they considered land rights to be an obstacle to development.
The present administration is implementing several actions against corruption. With an average score above 2.5 – and with less than 25 per cent neutral or undecided responses – respondents perceived these actions to potentially have a positive impact. Such an attitude is fully consistent with the emphasis put by respondents on the very negative influence of corruption on development.
Another action that gathered approval among the survey respondents was the reform of the power sector and the likely unbundling of the activity of the state monopoly in this area, Société Béninoise d’Energie Electrique. This, again, is in agreement with the negative opinion of respondents about the management of state-owned companies.
7 Response Heterogeneity
To complete the analysis, we now examine whether average scores in the population of respondents hide strong differences across specific groups, in which case the conclusions just obtained should be somewhat qualified. The way to proceed is simple. It consists of testing the statistical difference between the answers of different groups of respondents. To be consistent with the strategy used earlier, the emphasis is put on those cases where strengths and weaknesses, as defined by an average score, respectively, above 2.5 or below 1.5, are present in particular groups of respondents but disappear when considering the whole population. This analysis is performed on three subgroups: women vs men, formal firm managers vs other respondents, and financial managers vs other respondents.
Table 3.4 illustrates the procedure for the women/men dichotomy (for results on other subgroups se Houssa and Bourguignon, Reference Houssa, Bourguignon, Bourguignon, Houssa, Platteau and Reding2019). Questions appearing there are ranked according to the degree of statistical significance of the difference in average scores between the two groups. Two situations arise. The first case is where men are strongly positive in their answer and women much less so, so that the general average scores are in the neutral interval (1.5, 2.5). This is the case for the confidence that men seem to have in political institutions like the Supreme Court or in the discussion of the budget in the parliament. The other case is women being strongly negative but men being neutral, so that, again, the overall average score is in the neutral interval. This occurs for the question on the autonomy of trade unions, for instance, or the question on the constitutionality of some government actions. Of course, there are also cases where the difference between men and women is significant but on the same side, so that the overall average score is little affected. This is the case for the question on familiarity with land laws, for which both men and women were negative but to varying degrees.
Table 3.4 Top issues with significant differences between men and women
| Question | Average score for women | Average score for men | T-stat |
|---|---|---|---|
| How truthfully and seriously is economic policy (e.g. fiscal policy, taxation, trade, etc.) debated within the government and in parliament? | 1.97 | 2.73 | 3.87 |
| To what extent do you think that the Haute Cour de Justice, the Constitutional Court, and the Supreme Court effectively enforce compliance with the formal rules of the constitution? | 1.53 | 2.57 | 3.73 |
| How reliable (in terms of realism, consistency, coverage, degree of detail, coherence) is the budget? | 1.35 | 2.19 | 3.28 |
| To what extent do you share the view that foreign aid improves the quality of economic policy | 0.96 | 1.74 | 3.22 |
| To what extent are trade unions autonomous vis-à-vis majority political parties? | 0.83 | 2 | 3.16 |
| How familiar are you with Beninese land law, i.e. the Land Acts? | 0.92 | 1.5 | 3.13 |
| To what extent do parliament and the executive function according to the constitution? | 1.32 | 2.21 | 2.97 |
The question does arise as to why opinions may differ between men and women on such crucial issues as whether political actors behave according to constitutional rules. A possible explanation is that many women in the sample of respondents operate in the informal sector of the economy and may not have the same familiarity with this kind of issue. Also, they may not have the same level of education as other respondents. If this explanation is correct, then the constitutionality of political action in Benin should be added to the list of the country’s institutional strengths.
The same analysis with respect to formal firm or financial institution managers also reveals clear differences in information sets. For instance, financial managers had more concerns than other respondents about issues related to land and involving formal companies in urban areas, or about the ability of the judiciary system to resolve corruption problems. Not surprisingly, they were more satisfied with banking regulation. Formal firm managers, on their side, were more sensitive to the lack of government transparency on subjects related to economic policies and the budget. Except for this, differences with other respondents were more a matter of intensity than direction, the same being true of financial managers.
C The Main Lessons from the Opinion Survey: Summary
Although corruption cannot be considered an ‘institutional area’, it clearly appears in the opinion of the respondents as a major cause of institutional weaknesses across the board. It is certainly behind the low opinion expressed in the survey about public management, the dissatisfaction with the business environment, and the doubts expressed about the functioning of the political system. Corruption is felt to be present everywhere in the economic and political system.
This emphasis on corruption illustrates the fact that the kind of opinion survey undertaken for this study of Benin’s institutions, like the international comparisons based on synthetic indicators in Section I of this chapter, provides more information on what people and experts feel works well or not so well, than on the dysfunctions or the positive role of institutions per se. Corruption is certainly a plague in Benin, but its effects are not necessarily well identified.
Concerning the broad institutional areas, public administration is found to be the weakest link in the functioning of the Beninese economy and society, without it being completely clear what does not work there, except for the deleterious effect of corruption. For instance, no strong opinion was expressed on civil servants – except for their frequent strikes – or the organisation of the whole sector. What is clear, however, is the way this perceived weakness of the public administration is behind the dissatisfaction with the business environment, which is another strong message of the survey. As far as political institutions are concerned, answers to the questionnaire show some ambivalence, with respondents expressing some confidence in the way the system works and in current reforms, while at the same time, here again, pointing to the harm done by corruption.
Other weaknesses stressed by the opinion survey include the lack of transparency with regard to state actions. This may be the reason why no clear view about the state’s dysfunctions was expressed in the survey responses. Opacity makes evaluation difficult, except perhaps when results are directly apparent, as is the case with state-owned companies – in the power sector in particular.
On the positive side, there was broad agreement on civil liberties and the state being free of the influence of religion or traditional culture. Such circumstances doubtlessly should be favourable to private initiative and unbiased policy making. Yet, there is some lack of consistency here between this perceived autonomy of the state, on the one hand, and the sense of the detrimental effect of corruption, on the other.
Overall, the expert opinion survey is a bit disappointing in the sense that it does not point to well-defined obstacles to development arising from the working of institutions in Benin. A possible reason for this may lie in the heterogeneity of opinions depending on where respondents stand in the workings of the economic and political system. This is apparent when comparing the answers of formal firm or financial organisation managers and those of other respondents. Strong perceptions in opposite directions by different groups of respondents may tend to neutralise each other. Table 3.A.1 in the Appendix to this chapter illustrates that heterogeneity by showing how the direct choice of critical broad institutional areas in the first stage of the survey differed across selected groups of respondents.
III Institutional Implications of ‘Growth Diagnostics’ and Similar Exercises
To end this review of insights into the way institutions in Benin may create obstacles to the country’s development, we now briefly review the potential institutional implications of ‘growth diagnostics’ exercises that have been conducted in Benin in the spirit of the Hausmann et al. (Reference Hausmann, Rodrik and Velasco2005) methodology over recent years.
Two studies of this type have been completed over the last ten years or so: the first by Ianchovichina (Reference Ianchovichina2009) for the World Bank and the second for the International Monetary Fund by Barhoumi et al. (Reference Barhoumi, Cui, Dieterich, End, Ghilardi, Raabe and Sola2016). The former is rather complete but a bit old, as it essentially refers to the period 1996–2006. The latter is more recent, but less complete. A related report was released more recently by the World Bank (2017b); this presented a Systematic Country Diagnostic for Benin built upon a different methodology than growth diagnostics. We summarise the main findings of these studies in the following paragraphs, insisting on the points that are directly related to the working of institutions. We also complement them with some of the results of the World Bank (2009, 2016) Enterprise Surveys based on a sample of firms operating in Benin, as these provide further interesting evidence on some of the points raised in the preceding studies.
A The World Bank 2008 Growth Diagnostic
The growth diagnostic approach relies on a simple model of optimal growth leading to a set of key determinants of growth performance. Considering these determinants one by one, the objective is then to determine the extent to which they are constraining the development of a country in a given time period.
Referring to the decade ending in the mid-2000s, Ianchovichina (Reference Ianchovichina2009) identifies three sets of binding constraints.
1 Poor Quality of Infrastructure
As of 2008, Benin displayed infrastructure deficiencies in different areas. Notably, power supply (in quantity and quality) was the leading constraint on business, as most firms had to bear the cost of installing their own power-generation capacity. In the same way, poor services in railway and roads undermined Benin’s geographical advantage to serve landlocked countries (Burkina Faso and Niger) to its north. Moreover, lack of adequate rural roads, poor logistics in transport and storage facilities, as well as deficiencies in water management and irrigation impeded progress in agriculture and the agrobusiness industry.
2 High Risks on Return Appropriation: The Tax Issue
In the Investment Climate Assessment of 2004, used in the World Bank growth diagnostic, firms reported difficult challenges in dealing with the tax administration: a complex tax system coupled with high tax rates, a heavy bureaucratic burden, and corruption. In the same way, they reported serious problems in the judicial administration: long and costly litigation procedures in resolving conflicts, especially in land and financial markets, and, there too, a high level of corruption. These were considered as strong deterrents to business dynamism.
3 Poor Quality of Human Capital
Although the availability of skilled labour did not appear to be a binding constraint in 2008, it was noted that Benin was lagging in terms of the quality of education, so that it was felt that human capital could become a constraint in the future. This is still the case today. A recent comparative analysis among ten francophone African countriesFootnote 10 shows that primary school pupils in Benin are performing worse in reading and mathematics than those in peer countries (PASEC, 2014).Footnote 11 Moreover, significantly low learning competencies are found for children in rural areas, as well as those from poor families. This limits inclusive growth and has certainly contributed to the rise of inequality over the past years.
The World Bank 2008 growth diagnostic also noted that, by 2005, the pressure on land was mounting. If the utilisation rate of land was still well below full capacity in the north of the country, this was not the case in the south. For instance, the utilisation rate of cultivable land in the département of Ouémé was reported to be 96 per cent.
As can be seen, several of these binding – or potentially binding – constraints identified in 2008 are related to institutional issues that have been mentioned in the opinion survey completed for the present study.
B The 2016 IMF Growth Diagnostic
The Barhoumi et al. (Reference Barhoumi, Cui, Dieterich, End, Ghilardi, Raabe and Sola2016) study, completed ten years later, is not as comprehensive. It focuses on the way investment may be scaled up in Benin. The binding constraints that it identifies echo those identified by Ianchovichina (Reference Ianchovichina2009) and the opinion survey analysed earlier in this chapter. Of special importance in that study are the infrastructure constraint, especially in the power sector, and the tax system, which is seen as being responsible for lower tax revenues and therefore an impediment to the scaling-up of investment. Concerning the tax system, the diagnostic insists both on the complexity of the system, but also on the inefficiency of the tax collection apparatus, which leads to many firms simply not paying taxes, either legally through loopholes or illegally through corrupt practices. The reason why the tax/GDP ratio of Benin is comparable to that in other sub-Saharan African countries is essentially because of the relative importance of customs duties on re-exports in the direction of Nigeria.
C The World Bank 2017 Systematic Country Diagnostic
The Systematic Country Diagnostic (World Bank, 2017b) replaced the old Country Assistance Strategy documents in the relationship between the World Bank and low-income countries. It is the analytical background document for the preparation of the Country Partnership Framework (CPF). In the case of Benin, the last CPF was signed in 2018, for the 2019–2023 period.
The Systematic Country Diagnostic 2017 identified the following areas of weakness for the development of Benin, and therefore pathways of action within the CPF: infrastructure, with emphasis this time on transport and logistics in order to capitalise on the port of Cotonou; informality, caused by the illegal nature of cross-border trade with Nigeria; service delivery, especially in the education sector; and the need for developing more effective social safety nets. In addition, the Government of Benin committed in the final version of the CPF to enhancing its efforts in improving public management, and governance more generally.Footnote 12
Again, several of these areas match some of the conclusions derived from the opinion survey carried out for the present study, especially those concerned with public management, infrastructure, and, implicitly, corruption, since this is what is behind the commitment to better ‘governance’. From that point of view the Systematic Country Diagnostic and CPF are quite clear – it is said in the opening remarks:
The political economy [of Benin] is characterised by a concentration of powerful interests and a resulting uneven playing field, weak institutions, poor governance, and incidents of corruption. As elaborated in the SCD [Systematic Country Diagnostic], Benin’s potential for achieving the twin goals [i.e. poverty reduction and shared prosperity] has faltered for several reasons, including those related to political economy: low levels of trust between economic agents, weak institutions, and poor governance.
If such an official document, endorsed by the government, is so clear, it may be surprising that the respondents to the survey analysed in the preceding section were shyer in their evaluation of Benin’s institutions. The reason has probably to be found in the mechanical format of the questionnaire, which in some cases did not allow respondents to express their deep convictions.
D The 2009–2016 World Bank Enterprise Surveys
Although there was no diagnostic attached to them, it seems interesting to check the Enterprise Surveys carried out by the World Bank, to see whether their findings match the binding constraints identified by the preceding growth diagnostics. The answer is that they do. In the various types of information collected by these surveys, the largest differences between Beninese firms and firms in other sub-Saharan African countries appear under the following headings:
Corruption: bribery incidence declined between 2009 and 2016 and is lower in Benin than in other sub-Saharan African countries, but Benin very much dominates other countries in terms of gifts given to get government contracts, construction permits, or a favourable judgement in court.
Infrastructure: power supply is much lower and outages are more frequent in Benin than in other sub-Saharan African countries; moreover, the situation has been getting worse since 2009.
Informality: seen as a major source of unfair competition by formal firms, again more in Benin than in the rest of sub-Saharan African. The same applies to tax rates and the tax administration.
Summing up, the growth diagnostic exercises conducted in relation to Benin over the last decade or so are rather convergent in pointing to several key weaknesses that have clear institutional roots: intense corruption, inefficient public management (including infrastructure, service delivery and, especially, the tax administration), and a high level of informality.
IV Conclusion
This chapter reviewed expert opinions on the quality of institutions in Benin and the way this could affect the country’s development performance. Three types of evidence were considered: synthetic indicators available in cross-country databases; a specific opinion survey carried out among local decision makers of different types and engaged in different activities; and analysis of the institutional implications of binding economic constraints identified in several recent growth diagnostic exercises. These various sources converge in pointing out several institutional weaknesses that impede the acceleration of development in Benin, even though they may not always agree on the severity of these institutional constraints.
Corruption is unanimously seen as the most serious impediment to the good functioning of institutions and a favourable development context. Corruption is found to affect practically all sectors of the economy at all levels of responsibility. This is recognised by both the respondents to the opinion survey and the authors of growth diagnostic exercises. Comparison with other countries in the region or countries that have outperformed Benin over the last decades is less conclusive. If Enterprise Surveys find that, from the point of view of business, the situation in Benin is substantially worse than in the average sub-Saharan African country, country-by-country comparison leads to different conclusions. The degree of corruption in Benin, as can be appraised through synthetic indicators, turns out to be roughly comparable to that in neighbouring countries. Corruption might be even less serious than in several countries that grew faster than Benin over the last twenty years, this being true today as well as ten or twenty years ago. Such findings may reflect the conceptual imprecision of synthetic corruption indicators, but they also call for a more nuanced analysis of the effects of corruption on the development of a specific country.
Weak public management is the second unanimously recognised source of hindrance in the process of development. Of course, this may partly be the consequence of corruption. Here too, the cross-country difference in synthetic indicators of the quality of public management across countries is not strongly unfavourable to Benin. Yet, some sectors are singled out as particularly weak by survey respondents and analysts. Three of them are repeatedly singled out. The tax system is found to be complex and the tax administration grossly inefficient in collecting tax revenues, with clear adverse consequences for the dependency of Benin on foreign finance. The power sector, run by a state-owned monopoly, is found to perform badly due to weak or ineffective regulation. Finally, if the delivery of social services, especially education, is found to have made progress in quantity, this is not the case for quality. Benin underperforms in relation to other sub-Saharan African countries by a wide margin and, from that point of view, lags very much behind the countries that grew faster, from roughly the same initial level of income, over the last twenty years.
The opacity of government policy making to the public, very much stressed by survey respondents, is probably to be imputed to weak public management, but it is also a sign of deficient political institutions, generally regarded as weaker than in other sub-Saharan African countries. From that point of view, however, survey respondents are somewhat ambivalent. On the one hand, many of them tend to trust constitutional institutions and are confident of the success of some current reforms. On the other hand, most also agree that the whole system is deeply corrupt and, because of this, often dysfunctional. Such a severe judgement even appears in the opening remarks of the official CPF, a document signed between the Government of Benin and the World Bank.
Available statistics show that informality is more developed in Benin than in the average sub-Saharan African country. Growth diagnostic analyses suggest that informality has a cost in terms of tax revenues, job precariousness, and lack of control over the economy. This is not a point that appears strongly in the opinion survey, perhaps because of the presence of a substantial group of informal firm managers in the sample. It is not a dimension of institutions that appears explicitly in the synthetic indicators provided by international databases. Yet, the reason why informality is more developed in Benin is clear: it is more the result of the importance of the illegal cross-border trade with Nigeria than it is the result of some specific institutional failure. However, its consequences for the functioning of institutions are serious.
A last area deserves mention, even though it was not prominent as such in the opinion survey and was not explicitly covered by the synthetic indicators: it is the way land allocation is managed. One of the growth diagnostic studies mentions that land is becoming scarce in the southern part of the country, so that managing it efficiently will become more and more crucial in the future. As in other sub-Saharan African countries, land operations raise difficulties in Benin because of the uncertain status of ownership and the legacy of customary practices. A reform was passed in 2013 that, according to the opinion survey, is complex and does not really resolve the sources of land conflict. Land laws and their implementation reveal institutional weaknesses whose economic consequences may be considerable in the future, especially in a country with a comparative advantage in agriculture.
Open access