Sector by Sector Towards Feasible Anti-Corruption Strategies for

Antonio Andreoni Senior Lecturer in , SOAS University of London & SOAS ACE Research Director Visiting Associate Professor, SARChi Industrial Development, University of Johannesburg Tanzania at a crossroads: economic growth

Source: 2017 • Tanzania has achieved significant growth performances since 2005

10

8

6

4

2

0

-2

-4 Tanzania Advanced economies Emerging market and developing economies World Tanzania at a crossroads: population booming

> 90 millions by 2030 (100 millions by 2035) Tanzania at a crossroads: governance

• despite poor scores on its governance indicators (especially with respect to the control of corruption and government effectiveness)

Control of corruption Government effectiveness

80.00 80.00 70.00 70.00 60.00 60.00 50.00 50.00 40.00 40.00 30.00 30.00 20.00 20.00 10.00 10.00 Global PercentileRanking of Country Global Percentile Ranking ofRanking Country Percentile Global 0.00 0.00 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

TANZANIA KENYA SOUTH TANZANIA KENYA Tanzania at a crossroads: structural transformation?

• without sustained structural transformation - agriculture still dominant (1/4 GDP, 2/3 Workforce), low productivity and constrained by capabilities/productivity/governance bottlenecks

- upward trend in manufacturing but discontinuous, contributing less than 10% to GDP, slow and limited changes in industrial competitiveness and export performances, dualistic structure of the local production system

- Informal service economy dominant in urban areas and often linked to rents capture activities Tanzania at a crossroads: sectoral trends

Source: World Bank 2017

2.5E+10 55

50

45 Manufacturing, value added 2E+10 (constant 2010 US$)

40 Industry, value added (constant 2010 US$) 35 Agriculture, value added (constant 1.5E+10 2010 US$) 30 Services, etc., value added (constant 2010 US$) 25 Agriculture, value added (% of GDP) 1E+10 20 Manufacturing, value added (% of GDP) 15 Industry, value added (% of GDP) 5E+09 10 Services, etc., value added (% of GDP) 5

0 0 Tanzania at a crossroads: low productivity

Source: World Bank 2017 • Agriculture (68% employment), low VA per worker & (e.g. fertilizers; irrigation; etc.) 600 10

9

500 8

7 400

6

300 5

4

200 3

2 100

1

0 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Fertilizer consumption (kilograms per hectare of ) Agriculture value added per worker (constant 2010 US$) Tanzania at a crossroads: dualistic structure

Source: Andreoni 2017 Total export by indusrial sectors and establishment MVA by industrial sector and major size manufacturing industries and by establishment size 3,000,000 4,500,000

4,000,000 2,500,000 3,500,000

2,000,000 3,000,000

2,500,000 1,500,000 2,000,000

1,000,000 1,500,000

1,000,000 500,000 500,000 0 0 1-4 5-9 10-19 20-49 50-99 100-499 500+

Manufacture of rubber and plastics products Manufacture of food products Manufacture of other non-metallic products Manufacture of products Manufacture of tobacco products Manufacturing Manufacture of beverages Mining and quarrying Manufacture of food products Total exports Manufacturing Manufacture of rubber and plastics products Mining and Quarrying Manufacture of other non-metallic mineral products Manufacture of textiles 80%VA, 200 est 87%VA, 200 est Window of opportunity, and recent challenges…

Tanzania faces a window of opportunity for addressing corruption in the country and unlock its growth potential towards structural transformation 1. A political settlement in flux: Dramatic acceleration in the fight against corruption combined with pragmatic and deal-making approach of the new leadership, is opening new forms of dialogue between the public and private sectors and potentially reshaping the growth-governance relationship. 2. Combination of vertical and horizontal anti-corruption strategies: For recent progress in anti-corruption to be sustainable, sector-specific and incremental anti-corruption reforms need to be embedded in institutions and target specific processes, deliver pragmatic solutions and tangible results (quickly, risk of running out of time towards 2019 turning point), while opening new spaces for productive investments and diversification. 3. Structural transformation for sustained and inclusive growth: The achievement of these development outcomes in anticorruption is a critical ingredient in sustaining Tanzanian efforts towards its structural transformation 2015 Election: Taking back control of corruption

CCM Magufuli (58.5%) The evolution of the political settlement: the long term view and some recent developments

• The Nyerere developmental state and the deep roots of clientelistic networks in Tanzania (1961-1985)

• Privatisation, multi-partitism and corruption: the weakening of the dominant party under Mwinyi (1985-1995)

• The Mkapa reforms and the response to corruption of a weak dominant party (1995- 2005)

• Grand corruption and competitive clientelism under Kikwete (2005-2015)

• The “bulldozer” or the “builder”? The vulnerability of the authoritarian coalition and the potential developmental state under Magufuli A political settlement in flux: bulldozer or builder? • Bulldozing the political settlement: The first phase (until late 2016) of the new government was characterised by: - a strong move to centralise power around the President; - A confrontational and authoritarian approach towards the private sector; - an increasing pressure on the opposition and media. •This approach suggests a political settlement characterised by an authoritarian government that is vulnerable to losing power, and seeks to strengthen its control over its own supporters (internal political factions and “regional lords”) and opposition by • breaking consolidated clientelistic networks • using repression of the media and opposition’s political space (in particular its regional rooting – CCM comparative advantage) A political settlement in flux: bulldozer or builder? • Building a new political settlement? More recently a shift towards: - institutionalisation of power within CCM, and coalition building; - disciplining the corrupt distribution of resources with a “pragmatic approach”, including direct involvement in deal-making and investments (although without articulated policy framework); - continued systematic repression of the oppositions and the media. •A subtly different political settlement, one where • authoritarianism is balanced by weakened opposition and strong top-down control over the party and government apparatus • government attempts to re-allocate rents and establish new terms of engagement and procurement (mining, power, infrastructure…) and different private sector responses (new forms of dialogue, uncertainty, …) Deal-making approach: cases

• In October 2016 inauguration of a $120 million fruit processing factory in the outskirts of (Bakhresa) • In December 2016 deal with Dangote which led to the re-opening of the biggest cement factory in the country • In January 2017, concessional loan (from ) for the construction of a 400 kilometre-stretch of the envisaged 1,200 kilometre standard-gauge central railway • 2018 Sugar licencing allocation to 4 major sugar producers

…Other cases

• Signalling willingness to allocate rents, but also discipline them • Requesting businesses who cumulated capital (also thanks to their political connections / rents capture) to invest them in new industries The evolution of the political settlement: forecasting political settlement trajectories in 2017 What anti-corruption approach?

So far the anti-corruption agenda has relied mainly on vertical efforts: • Rule by fear: blaming and shaming allegedly corrupt people in both the public and private sector and direct involvement in anti-corruption operations, backed by the police and the intelligence. • Punishment: removing people from key positions in public offices and authorities, and prosecution enforcement (also in the form of ultimatums and amnesty agreements). • Centralisation: reducing potential resource leakages from the centre to the regions/districts by direct allocation of resources (e.g. fertilizers, education) and centralisation in the President’s Office of resource allocation functions (e.g. sugar import licence)

> Vertical forms of enforcement of the rule of law While the evidence is still scattered, vertical anti-corruption strategies may have achieved some results. Is this anti-corruption approach sustainable?

However, there are reasons for concern about the sustainability of vertical strategies and their effectiveness in terms of outcomes.

• Vertical anti-corruption efforts are mainly attacking the manifestations of corruption created (and enabled) by a much more diffuse and diversified set of processes, unfolding in different sectors. Opportunity of horizontal anti-corruption strategies for Tanzania • Combining vertical and horizontal anti-corruption strategies (ACE Approach) increases the chances of delivering better development outcomes and make anti-corruption efforts more sustainable (avoidance of resistance/backfire) because: - take into account the existing political settlement (feasibility) - target incremental changes by addressing different and specific types of corruption processes in each sector - provide powerful organisations in the sector with alternative ways to generate income or coordinate activities in their own interest (e.g. open new opportunities for productive investments in the private sector) beyond existing corruption practices • It ALSO offers an opportunity to shift from a risky vulnerable authoritarian political settlement to a more effective developmental state able to deliver development outcomes in a sustainable manner Sector by sector…

Resolving Designing for PORT & ENERGY rights LAND CUSTOMS SECTOR differences

EPZA & HEALTH SKILLS MEDIA incentives

SUN Aligning FLOWER TO PHARMA Building incentives OIL TEXTILE coalitions Employment promotion challenges in African SKILLS countries

• Supply/institutional problem: • Limited availability of a skilled workforce, especially vocational skills • Limited presence of high quality VET institutions/programmes (although some pockets of excellence), and political economy challenges in the skills sector (along all the chain, vulnerability to rents capture and lack of accountability) • Demand/firm level problem: • Limited number of competitive firms capable of organizing and deploying skills effectively, thus limited impact of skills on productivity and limited incentive in hiring skilled workers (and wages premium) • Limited number of competitive firms able to identify the specific type of skills they require to adopt new production models and technologies • Demand/economic structure problem: • The production base of the economy is small (few companies) and disarticulated (few local supply chains and over dependence on imports – underdeveloped LPS) • The industrial structure is also dualist – i.e. a few big, many small and very few medium size firms (so called missing middle) Labour force employment and inactivity SKILLS trends 2007-2016: Tanzania and East Africa

Labour force employment (Tanzania and East Africa) Inactivity trends in Tanzania (male and female)

Source: LO/FTF 2016, ILO data Specific and interrelated political economy issues SKILLS (and cumulative dynamics): what feasible strategies?

Sectoral/cross-sectoral dynamics Supply/institutional problem > Political economy of skills development Country/region dynamics Demand/economic structure problem > Political settlement of the country (and Demand/firm level problem > region – EAC) Political economy of capabilities and competitiveness development Supply/institutional problem: SKILLS The political economy of skills development in Tanzania

• Financing models • Skills levies on firms, between 1-2% (Kenya, South Africa, Tanzania 4.5%, …) • Financing burden and unintended consequences (formal employment disincentives, fixed costs, frequency, difficult especially for SMEs…) • System vulnerabilities from collection to allocation and use (rents capture opportunities along the chain involving firms, Pb & Pv providers, regional boards) • Institutional setting and Governance models • Authorities (often) operating as both regulators and provider and limited accountability of providers (public VET distributed, private providers) • National VET systems are fluid/diverse: numerous informal practices with mixed effects • Given limited accountability, private-public sector mistrust and firms ‘exit’ option • Outcome and quality • Double goals and trade offs: youth inclusion and sector specific high quality training • Resistances in curricula reforms and emphasis on certification more than employability among VET institutions • Difficulties in involving private companies (perceived as double burden and cost/risk e.g. Uganda) • Limited accountability (and ‘voice’) of the users Supply/institutional problem: Incentives structure around financing

SDL as General Domestic Revenue and Large Taxpayers Revenue 300000.00 80%

70% 250000.00

60%

200000.00 50% Total SDL (A1+C1)

150000.00 40% SDL General Direct Tax Contribution 30% 100000.00 SDL Large Taxpayers Direct 20% Tax Contribution 50000.00 10%

0.00 0% TOTAL TOTAL TOTAL TOTAL TOTAL TOTAL TOTAL TOTAL 2010-11 2011-12 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 (2Q)

Source: Andreoni, 2018, based on TRA data Supply/institutional problem: Incentives structure around financing

300000.00

Total SDL (A1+C1) 250000.00

1/3 TOTAL SDL

200000.00 1/3 VETA (source: Auditor General)

150000.00 VETA Recurrent Budget Allocation (source: National Budget 2017) 100000.00 HESLB (SDL Quota self reported is between 60 and 70)

50000.00 Skills Development Programme (Ministry Labour)

0.00 2010-11 2011-12 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 (2Q)

Source: Andreoni, 2018, based on TRA data and other sources Supply/institutional problem: Incentive structure around quality & outcome

Training types (courses) Long Short VETA-owned Average Ratios 53% 47% Private VET Average Ratios 80% 20%

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

Long Courses Short Courses

Source: Andreoni, 2018 Demand/firm level problem: The political economy of capabilities and SKILLS competitiveness development • Majority of firms tend to be politically (not production) competitive (yet) • Political connections and rents capture allow companies (and sometimes conglomerates) to operate in the economy (specific sectoral and value chain specialisation) • Rents opportunities more than production/market opportunities guide firms investments, as a result limited organisational capabilities for enhancing productivity and deploying skilled labour effectively • Dual industrial structure: A few firms control entire sectors and export activities • Challenges in involving companies, despite various strategies and some successes • Rebate systems (Kenya, Uganda, South Africa, not in Tanzania fully/yet) • Identification of firms for on the job training and employability incentives is challenging • Risk of opportunistic engagements in skills programmes with limited employability results • Sectoral councils (difficult with few companies, risk of capture and limited additionality) • Sector-ringfenced funding (South Africa), but risk of negative impact on nascent sectors • PPP models have been successful in some cases, but difficult to scale up (e.g. IMTT Moshi Centre in Tanzania) • Opportunities for sector-specific training and cross-sectoral spillovers • Limited offer of combined support of companies in organisational capabilities development (agro-mfg extension services, intermediate technology institutions, ….) ACE Feasible Strategies in the skills sector in Tanzania: SKILLS Realigning incentives and designing for differences

Understanding the sectoral incentive structure

Discreet Choice Experiment (DCE) and Firm Level (300+ Stratified Sample) Capability Assessment Tailoring feasible strategies for different (ongoing in Tanzania) types of firms PORT & Port and smuggling: Why? What impact? CUSTOMS

• The centrality of the Dar es Salaam Port for the Tanzanian economy (of corruption) • Port/Import licensing as critical in the political economy of Tanzania • Department for Disaster Management /PMO: the management of pricing and scarcity • Import licensing and export bans as KEY political tools in distorting the internal (across regions) and regional markets across EAC custom union/common market (e.g. bans of grains exports) • CET revisions (SI and exemptions) and the political economy of dumping – within the URT or an external player? • sugar and in particular, with large constituencies involved along sectoral value chains, from farmers to producers, to distributors/traders. PORT & Port and smuggling: Why? What impact? CUSTOMS

• The centrality of the Dar es Salaam Port for the Tanzanian economy (of corruption) • Port/Smuggling as critical for rents capture: businesses developed sophisticated ways to • Misinvoicing • Trade/smuggling collusion • Manipulating EAC CET (industrial sugar for domestic use; Pakistan-Kenya bilateral on rice) • Distorting procurement (ships waiting for a call, delivering in 2 days from Brazil!) • Smuggling affecting domestic production negatively • strong disincentive to invest • unfair price practices and farmers squeezing • Limited production capacity utilization among domestic producers with an impact on productivity and price competitiveness • De-industrialization (factories transformed in warehouses) Smuggling routes: PORT & international, regional (EAC) and national CUSTOMS

Smuggling: Smuggling: - Direct via Zanzibar to - Direct Dar Dar - Indirect EAC - Direct via Zanzibar to - Transit small ports - Indirect via Zanzibar PORT & Sugar underreporting (mirror statistics) CUSTOMS

Key countries accounting for underreporting are: 2.00e+07 , United Arab

1.00e+07 Emirates, Thailand, , Brazil, Morocco, Egypt, Algeria 0

Trade in sugar between those countries and TZN amounts to 71%. -1.00e+07 -2.00e+07 Difference Tanzania vs foreign country in USD -3.00e+07 1995 2000 2005 2010 2015 Year About the Anti-Corruption Evidence Research Consortium Funded by UK , ACE is a partnership of highly experienced research and policy institutes, led by SOAS, University of London: • London School of Hygiene and Tropical Medicine (LSHTM) • Palladium, London • University of Birmingham • University of Columbia, New York • BRAC Institute of Governance and Development (BIGD), Dhaka • Transparency International Bangladesh (TIB), Dhaka • James P. Grant School of Public Health BRAC University (JPGSPH), Dhaka • Research on Poverty Alleviation (REPOA), Dar es Salaam • Economic and Social Research Foundation (ESRF), Dar es Salaam • Ifakara Health Institute (IHI), Dar es Salaam • Centre for Democracy and Development (CDD), Abuja • Health Policy Research Group, University of Nigeria

www.ace.soas.ac.uk @ACE_soas The evolution of the political settlement: 2018 and the approaching of the new elections… fighting internal contestation and delivering development outcomes