Reforming Tanzania's tea sector: a story of success?

An earlier version of this paper (which also contains an extensive discussion of the world tea market) has been published in the World Bank's Africa Working Paper series (Baffes, 2004a).

Abstract

Tea, one of Tanzania's major export crops, contributes about $30 million to the country's export earnings and provides employment to some 50 000 families. Despite the sector's early success, nationalisation of two estates along with neglect of the smallholder sector made it clear that only broad-based policy reforms would revive the sector. Reforms in the tea sector started much earlier than reforms in other export crop sectors. Furthermore, they were undertaken, and hence owned, by the government and the industry. The reforms have been by and large successful. There has been considerable supply response, tea quality has improved, and the research system has been very successful in developing and disseminating useful research findings to both estates and smallholders. However, some issues must be addressed, namely excessive taxation, over-regulation, and the trade policy environment.

An earlier version of this paper (which also contains an extensive discussion of the world tea market) has been published in the World Bank's Africa Working Paper series (Baffes, 2004a).

1. Introduction

Many sub-Saharan Africa countries reformed their commodity sectors during the 1990s. In most cases these reforms meant eliminating or reducing the monopoly powers of the crop boards, which typically handled most marketing and trade aspects of the commodity sectors in question. By most accounts, the outcome of these reforms has been mixed (see, for example, Akiyama et al., 2003; Shepherd & Farolfi, 1999). The share of export prices received by producers increased and growers receive payments promptly. Reforms were often accompanied by considerable supply response and increased entrepreneurial activity. However, the quality of public services such as research and extension declined. The functioning of input markets deteriorated as soon as provision of credit was delinked from the crop boards' monopoly status.

The objective of this paper is to assess the policy reforms in the Tanzanian tea sector. The next section examines briefly the history of the tea industry in Tanzania, along with the causes of the sector's deteriorating performance. Section 3 examines the entire reform effort, including the privatisation of the two nationalised tea estates, the restructuring of the Tea Board, the revival of the research system, and finally the restructuring of the local tea blending and packing industry. The penultimate section assesses the recent performance of the industry by identifying policy impediments, including trade policy, taxation, and over-regulation. The last section concludes.

Numerous conclusions emerge from this assessment. First, reforms of the tea sector started much earlier than Tanzania's other export crop sectors (e.g. coffee, cotton, cashews). Second, there was no apparent ‘supply-driven’ assistance or push for reforms; by and large, reforms originated from within the country, accordingly characterised by a considerable degree of government and sector ownership. Third, the reforms have been largely successful, although much remains to be done. Finally, it has taken a considerable amount of time and resources for the reforms to be carried out.

2. A brief history of the tea industry in Tanzania

Tea was first planted in Tanzania in 1902 when German settlers introduced the crop to the Agricultural Research Stations in Amani and Rungwe. Commercial production began in 1926 in Usambaras and Njombe. In 1934 Tanzania produced 23 tons of made tea – about 4.8 kilograms of green tea leaf are required for one kilogram of made tea. Production increased considerably after World War II, when the British took over the tea plantations. By 1960, tea production exceeded 3 700 tons.

Before independence, tea was produced only on estates and all tea-related marketing and research aspects of the sector were handled by the Tanganyika Tea Board, renamed the Tea Board of Tanzania after independence. Estate production is common in many African and South Asian tea producing countries. Smallholder tea farming started in the early 1960s under the supervision of the Ministry of Agriculture. In 1968 the government initiated a full-fledged smallholder tea development programme, and all aspects of smallholder tea marketing and trade were turned over to the Tanzania Tea Authority, established by an Act of Parliament as a state-owned corporate body. The responsibilities of the Authority were summarised in its official pamphlet as follows (Tanzania Tea Authority, 1989):

[The Tanzania Tea Authority is] responsible for all aspects of smallholder tea development as well as for the functions previously exercised by the Tea Board of Tanzania. The authority is empowered to: promote, supervise and implement programmes for the development of the tea industry. This includes the supervision of planting, cultivation and harvesting of tea; inspect plantations and green leaf, negotiate agreements for leaf processing and organize the purchase and transport of green leaf. [The Tanzania Tea Authority] also takes part in the establishment, control and management of tea factories, control tea marketing, and acts as national marketing agent; and advising and make recommendations to the Minister on the development of the tea industry.

The Tea Authority promoted smallholder tea production, typically on plots of about a third of a hectare. Smallholder tea production in Tanzania was supported by the donor community, especially two World Bank operations: a $1.4 million loan in 1966 – part of an agricultural credit project – and a $7.1 million loan in 1972 (Singh et al., 1977; World Bank, 1971). Smallholder production increased considerably, accounting for about a quarter of Tanzania's tea production during the early 1980s and as much as 29 per cent in the 1985/86 season. Most of the smallholder tea leaf went to the eight Tea Authority-owned factories for processing, and the rest to factories owned by the estates.

Despite its apparent success, as early as the mid-1980s there were signs of trouble in the smallholder tea sector, and by the early 1990s smallholder tea production was falling rapidly. By the mid-1990s its share had dropped below 10 per cent and by 1998 it had fallen to 5 per cent, the lowest level since tea had been introduced as a smallholder crop (see Figure 1). Contributing to the decline were low prices and late payments by the Tea Authority, inefficient processing factories, inadequate use of inputs, rundown transport equipment and roads connecting farms to tea factories (feeder roads), and declining yields because of a failure to switch to high-yielding varieties. Deterioration in the smallholder sector is vividly summarised in a World Bank (1983) report (interestingly, the assessment was made in 1983 when the smallholder sector was at its peak, implying that while the sector appeared healthy the fundamentals were indeed poor):

Even in traditionally high grade producing areas … there is a problem of engineering standards, lack of spare parts, power failures, non-replacement of machinery and overloading. There have also been substantial delays in payments to smallholders, as a result of Tea Authority's precarious financial position. … Accounting records show total ‘over payments’ for green leaf to smallholders amount to Tsh3.6 million, implying falsified weight and/or payment records. Similarly, per kilogram costs attributed to Tea Authority-managed estate production were up to three times higher than the price paid to smallholders, again implying great inefficiencies if not falsification of records. (World Bank, 1983: 24, 87)

Estate production followed a largely independent path, with output growing considerably during the 1990s. By the 1995/96 season estate yields were ten times those of smallholder plots. The high yields reflected, to some extent, the vertical integration of estate production. Estates have their own transportation equipment and processing facilities and so are not dependent on public infrastructure. Most of the workforce lives in housing provided by the estates, which also provide medical services, schools, and other facilities. The estates, therefore, have access to a constant flow of high-quality labour.

Figure 1: Production of made tea (tons). Source: Tanzania Smallholders Tea Development Agency

When Tanzania nationalised most large companies during the 1960s and early 1970s, tea estates were exempted – with two exceptions. Nearly all estates were foreign owned, and there were fears that nationalisation would lead to retaliation by the London tea auction, which at the time handled all Tanzanian tea (and indeed, most of world's tea). However, two estates were taken into public ownership because they formed part of the operations of nationalised companies involved in other activities. The Kwamkoro estate, operated by Bird and Company (Africa), with 630 hectares under tea, was nationalised in 1967, and the Bulwa estate, operated by the Sikh Sawmills Company, with 680 hectares under tea, was bought by the government in 1971. The Kwamkoro estate was placed under the Sisal Corporation, later taken over by the Tanzanian Sisal Authority, and the Bulwa estate was placed under the Tanzanian Wood Industry Corporation. The Tea Authority took over the Bulwa estate in 1977 and the Kwamkoro estate in 1986.

Before nationalisation the two estates together produced more than 1 800 tons of made tea. Following nationalisation production began declining, to reach a low of 300 tons in the mid-1980s as almost half of the original 1 300 hectares were abandoned, while another 250 hectares were only partially planted. The reasons behind the deteriorating performance of the two nationalised estates were similar to those in the smallholder sector. Reports by the Ministry of Agriculture blamed labour shortages because of late payments of wages and poor housing conditions for workers, inadequate maintenance of feeder roads, underinvestment in factories, poorly maintained transport equipment, lack of fuel-wood owing to inadequate replanting of trees, and lack of credit (Faber, 1995).

Public research on tea also ran into major problems. Before independence, research for the East Africa tea producing region was conducted by the Tea Research Institute of East Africa. Following the collapse of the East African Community in 1977, the research programme was transferred to the Ministry of Agriculture and Cooperatives, but the programme was inadequately funded. The tea estates contracted with a UK University to operate a tea research unit, housed in one of the estates. Although the research focused on production systems with high input intensity, as practised in estate tea production, occasionally valuable research results were transferred to smallholders.

3. The road to recovery

Deterioration of the two nationalised estates, the poor performance of the smallholder sector, and the collapse of the research system clearly signalled that broad-based policy reforms were needed to revive the tea sector. The first step was the privatisation and rehabilitation of the two Tea Authority estates, which took place from 1988 to 1993. Restructuring the research system came next with the establishment of the Tea Research Steering Committee in 1988, which recommended creating an independent research organisation (which was not done until 1996).

More pressure for reforms came in 1994, when the Ministry of Agriculture and Cooperatives recommended privatising the Tea Authority factories to open the way for a more efficient collection and payment system, creating the Tanzanian Smallholder Tea Farmers Development Agency to promote smallholder production, reducing and simplifying the tax structure to attract investment, and establishing a tea auction. These recommendations complemented those made by the World Bank (1994) that same year: allow valuation of tea exports at the market rate of foreign exchange, privatise management and perhaps ownership of the Tea Authority factories, and transform the Tea Authority into a regulatory entity.

3.1 Rehabilitating the East Usambaras tea estates

The first step toward reform was privatisation of the two nationalised estates in the East Usambaras, a process that extended from 1988 to 1993 – see Faber (1995) for an extensive review of the rehabilitation process of these two tea estates. In view of the poor performance of the estates, the government considered three options: do nothing and let the estates and their factories close; let the estates continue producing at a loss with minimal renovation over the next ten years, by which time their residual value would be zero; or fully rehabilitate them. For political reasons, the third option was selected.

The Commonwealth Development Corporation, a statutory corporation of the British Government which invests in enterprises in developing countries, was invited to purchase a 60 per cent equity share in the estates, with the rest to be retained by the Tea Authority. They would form a joint venture to rehabilitate the estates. The incremental funding was estimated at £5.9 million.

Rehabilitation – long, difficult, expensive, and ultimately successful – took five years and cost £2.6 million more than expected (the total cost reached £8.5 million). The two new estates provide employment to thousands of workers and produced an average of more than 3 600 tons of made tea during 1998–2003.

The privatisation and rehabilitation of the two tea estates was important to the success of tea sector reform and offered a number of important lessons, as Faber (1995: 1346) succinctly explains:

Without a program of reforms, the rehabilitations will not succeed; without the rehabilitations, the country will lack the foreign exchange to sustain the program. … Perhaps the most important lesson of all is that continuity of commitment, a deep pocket and plenty of patience are likely to be required of those undertaking rehabilitation projects. Great rewards may ultimately be gained but they have to be struggled for, often for longer than originally anticipated.

3.2 Privatising the Tea Authority-owned factories

One of the key steps in the tea sector reform was the decision to privatise the tea factories. Of the six Tea Authority-owned tea factories put up for sale in 2000, four were in private hands as of November 2001 and three of them have begun renovating the facilities and paying farmers more promptly. Long delays in payment were the norm under Tea Authority ownership.

The four privatised factories are Katumba and Mwakaleli, which are now under Wakulima Tea Company and managed by Tanzania Tea Packers; Mponde Tea Factory, renamed the New Mponde Tea Factory; and Maruku Tea Factory, renamed the Kagera Tea Company. The two factories now undergoing privatisation are the Lupembe Tea Factory and the Dabaga Tea Factory (Mdee, 2001: 10–11). See Table 1 for the output of made tea by company for the 1996–2003 period.

Table 1: Production of made tea by company (tons), 1996–2003

	1996	1997	1998	1999	2000	2001	2002	2003
Tea Authority^1	2 341	1 879	1 423	1 327	1 904	—	—	—
Kagera Tea Co.	—	—	—	—	—	382	234	447
Tanzania Tea Packers	—	—	—	—	—	1 012	5 114	5 597
Dhow Mercentile Ltd	—	—	—	—	—	241	538	528
Usambara Tea Growers Association	—	—	—	—	—	811	1 300	1 127
Brooke Bond	7 326	10 157	9 267	9 273	9 516	10 342	10 400	12 071
Mufindi Tea Company	3 044	4 151	4 871	4 153	3 610	4 101	4 283	4 743
East Usambaras Tea Company	2 649	2 159	3 964	3 613	3 606	3 244	3 693	3 532
Tanwat^2	645	856	1 457	2 305	2 565	3 004	—	—
George Williamson	1 967	2 584	2 746	2 237	1 875	1 686	1 285	647
Bombay Burma Co-Op	365	448	600	477	525	565	625	647
Ralli/Balangai Estates	1 421	818	549	241	168	121	123	90
TOTAL	19 768	23 051	24 876	23 626	23 769	25 509	27 564	29 457
^1Split into the Kagera Tea Co., Tanzania Tea Packers, Dhow Mercentile Ltd and Usambara Tea Growers Association.
^2As of 2002, ownership of Tanwat changed to Tanzania Tea Packers.
Source: Tea Board of Tanzania up to 2000 and FO Licht, World Tea Monthly (June 2004, p 258) the rest.

3.3 Separating regulation and smallholder promotion

This divestment of commercial activities was just one step in the reform of the Tea Authority. Next came the separation of the regulatory and development functions of the Tea Authority. The Tea Act of 1997 established the Tea Board of Tanzania and the Tanzania Smallholder Tea Development Agency. Their functions were set out in the Tea Regulations of 1999 – the principal institutions in the Tanzanian tea sector and their roles are outlined in Table 2.

Table 2: Principal institutions involved in the Tanzanian tea sector

Institution	Entity	Main functions and responsibilities
Ministry of Agriculture and Food Security	Government	Supervises the sector. Acts as liaison between the sector and the legislature and provides legal and policy guidelines.
Tea Board of Tanzania	Statutory body	Created under the 1997 Tea Act and the subsequent tea regulations of 1999, it regulates the tea industry. Collects the 2.5 per cent cess for research and its own operating expenses. Issues production, import, blending, and packing licences. The private sector is represented on the Tea board.
Tanzania Smallholders Tea Development Agency	Statutory body	Created under the 1997 Tea Act and the subsequent tea regulations of 1999, it promotes the interests of the smallholders. Advises and makes recommendations to the Ministry of Agriculture and Food Security. Promotes and solicits financing for the five-year smallholders tea expansion programme. Supervises the privatisation and liquidation of the tea factories formerly owned by the Tea Authority.
Area Tea Growers Associations	Private sector	Established under the 1999 tea regulations, these associations are registered under company ordinance and promote smallholder interests by soliciting project finance from the five-year smallholders tea expansion programme. As of November 2001 there were associations in Usambara, Rungwe, Mufindi, Kagera, Lupembe.
Tea Association of Tanzania	Private sector	Established in 1943, it promotes the interests of the industry as well as influences legislative and policy decisions. Its membership consists of tea producers, packers and blenders. The Tanzania Smallholders Tea Development Agency is a member, representing smallholders. The Tea Association of Tanzania is a member of the East Africa Tea Trade Association.
Tea Research Institute of Tanzania	Private sector	Created in 1996 (commenced operations in 1998), it undertakes research on problems affecting the industry and develops technology to improve tea quality and production. It is a limited company with a 10-member board: estates (4), smallholders (2), government (2), growers (1), and the executive director. Its funding comes from the cess administered by the Board.
East Africa Stakeholders Tea Committee	Private sector	Addresses problems of the tea industry common to all East Africa tea producers, including intraregional trade issues.
East Africa Tea Trade Association	Private sector	Established in 1956, its objective is to promote the interests of tea trade in East Africa. It compiles and distributes tea price and trade statistics and facilitates the operations of the Mombasa Tea Auction. Its membership consists of 300 entities including tea producers, exporters, brokers, packers, and warehousers.
Mombasa Tea Auction	Private sector	Tea auction in Africa began in 1956 in Nairobi under the auspices of the East Africa Tea Trade Association. It moved to Mombasa in 1969. As of October 1992 transactions take place in US dollars. Currently, it is the world's leading tea auction, trading most of East Africa's tea, including Tanzania.
Source: Author's interviews; Ndunguru (2001); East Africa Tea Trade Association and Tanzania Smallholders Tea Development Agency.

The Tea Board is responsible for regulating tea cultivation and processing, licensing tea blenders and packers, licensing and controlling tea exports and imports, collecting statistics on the tea industry, and representing the government in international tea fora.

The Tanzania Smallholder Tea Development Agency is responsible for promoting and developing the smallholder tea sector, advising the Ministry of Agriculture and Food Security on tea industry matters, conducting research on tea, and providing finance to tea research and extension services. The Smallholder Tea Development Agency is a member of the Tea Association of Tanzania, where it represents the interests of smallholders. Although the Tea Association of Tanzania has been around since 1943, it began to take a much more active role in 1989 when it became a private entity. Its objectives are to promote and protect the interests of the tea industry in Tanzania, to influence government policy affecting tea, and to negotiate on behalf of the industry with government, the Tea Board, and trade unions.

3.4 Reviving research

Following the collapse of the East African Community in 1977, tea research was placed under the auspices of the Ministry of Agriculture and Cooperatives and was funded by the government. By the mid-1980s the research programme was in a state of collapse. The Tea Research Steering Committee, which was formed in 1988 to arrest the decline in research activities, recommended the creation of an independent research organisation, to be funded through industry levies.

The Tea Research Institute of Tanzania was thus established in July 1996 as a non-profit organisation (TRIT, 2000/01). In July 1997, the staff of the Ngwanzi Tea Research station, a privately funded organisation in the Southern Highlands, was officially incorporated into the Institute. In October 1998 a similar transfer of the government-owned Marikitanda Tea Research Station in the East Usambaras took place. That same year Cranfield University (located in the UK) was appointed as the managing agent of the Research Institute.

Currently, the Institute is managed by a ten-member board, with broad representation, including estates, smallholders, and the government. As a non-statutory body the Institute can use merit and performance criteria rather than seniority to determine the salaries and promotion paths of its researchers. Dissemination of research findings to estates and small tea growers is managed by the institute's Technology Transfer Unit. The Institute is funded by the industry, receiving 1.5 per cent on the net sale value of made tea. Although smallholders contribute only a small portion to the tea levy (because of their small share in total output), one-third of the institute's budget is earmarked for activities to benefit them.

3.5 Restructuring the local tea blending and packing industry

More than a fifth of Tanzania's tea output is consumed domestically. Before the tea industry was liberalised, Tanzania Tea Blenders, Ltd, a government-owned monopoly, handled tea blending, packing, and distribution for all domestic consumption. Its activities began with a precursor institution in 1970, when the tea packing factory Brooke Bond Oxo Tanzania merged with the Tanganyika Tea Company (which operated two tea estates) to form Brooke Bond Liebig Tanzania, Ltd. In 1974 the company's sales and marketing division was partly taken over by the Tea Authority and a new state-owned company, Tanzania Tea Blenders Ltd, was launched with 60 per cent shareholdings by the Tea Authority and 40 per cent by Brooke Bond Liebig Tanzania, Ltd. Recently Tanzania Tea Blenders has gone for privatisation. Brooke Bond Tanzania (Brooke Bond Liebig Tanzania's name after 1982) was invited to take full control of Tanzania Tea Blenders but it declined.

Following liberalisation, six more companies obtained blending and packing licences (three are in Dar es Salaam and the others in Mafinda, Tanga, and Moshi). Tanzania Tea Packers, Ltd, based in Mafinda, accounts for almost 70 per cent of the local market. To protect local blenders and packers, the Tea Board has imposed a ban on imports of made and packed tea.

4. Recent performance and constraints

The reforms have had a significant impact on the sector in a number of ways. On the production side, considerable supply response has taken place. For example, production of made tea rose from 20 000 tons in 1990 to 30 700 tons in 2004, a more than 50 per cent increase. By comparison, global tea supply during this 13-year period increased by 20 per cent. Interviews with government officials and traders at the Mombasa auction indicate that the quality of smallholders' tea has improved considerably as well. In 2004 (the last year with complete data) Tanzanian tea fetched almost the same price as Ugandan tea, commanded a 6 per cent premium over Malawian tea, and sold for 8 per cent less than Burundian tea and 19 per cent less than Kenyan tea (see Table 3).

Table 3: Prices at the Mombasa Tea Auction by origin, 1996–2004

	Tanzania	Kenya	Uganda	Malawi	Burundi	Rwanda	Average^1
US dollars per kilogram
1996	1.10	1.45	1.16	1.27	1.36	1.35	1.28
1997	1.70	2.03	1.80	1.63	—	1.96	1.82
1998	1.22	1.97	1.36	1.28	—	1.39	1.44
1999	1.18	1.86	1.29	1.06	1.65	1.52	1.43
2000	1.58	2.11	1.59	1.17	1.93	1.86	1.71
2001	1.22	1.62	1.12	1.06	1.27	1.47	1.29
2002	1.24	1.55	1.20	1.18	1.34	1.48	1.33
2003	1.25	1.62	1.25	1.14	1.43	1.56	1.38
2004	1.30	1.61	1.32	1.23	1.41	1.69	1.45
Tanzania's premium(+)/discount(−) (per cent) against other origins
1996	0	−24	−5	−13	−19	−19	−14
1997	0	−16	−6	+4	—	−13	−7
1998	0	−38	−10	−5	—	−12	−16
1999	0	−37	−9	+11	−28	−22	−17
2000	0	−25	−1	+35	−18	−15	−7
2001	0	−25	+9	+15	−4	−17	−6
2002	0	−20	+3	+5	−7	−16	−7
2003	0	−23	0	+10	−13	−20	−9
2004	0	−19	−2	+6	−8	−23	−9
Period average premium(+)/discount(−)
1996–99	0	−29	−7	−1	−24	−17	−13
2000–04	0	−22	+2	+14	−10	−18	−8
— not traded at the auction during that year.
^1Arithmetic average of the prices reported in the table.
Source: International Tea Committee.

Most of Tanzanian tea is exported. Two major importers are the UK (40 per cent) and Pakistan (15 per cent). Smaller importers include the US and Canada (6 per cent together) and Somalia, South Africa, and Sudan (3 per cent together). Most of the remaining tea goes through the Mombasa Tea Auction. (See Figure 2 for the marketing structure of the Tanzanian tea sector.) Comparing the 1996–99 period with the 2000–04 period shows that Tanzania's discount over Kenya's tea – the highest quality among the teas sold at Mombasa – declined by seven percentage points (from 29 to 22 per cent) while the corresponding average discount from all East African competitors declined by five percentage points (from 13 to 8 per cent).

Figure 2: Tea production and marketing structure in Tanzania. Notes: The numbers in parentheses are approximate production shares for the 1999/2000 season

The recently created Area Tea Growers Associations play an active role in representing the interests of the smallholders in a number of ways. For example, the Rungwe Small Tea Grower's Association – which represents about 15 000 smallholders – recently purchased approximately $4.5 million's worth of CaféDirect shares, UK's largest Fair-trade hot drinks company (The Public Ledger, 25 October 2004). According to the arrangement, CaféDirect will guarantee a minimum price to smallholders, regardless of the price fetched at the Mombasa auction.

Research is on a solid footing. The two research stations are working on several projects, ranging from the development of new tea clonal varieties to optimal use of fertiliser and soil and water conservation, while the technology transfer unit successfully disseminates research results to tea growers. The Research Institute is also engaging in contractual extension services with the newly privatised tea companies.

These are solid achievements, but several issues still require attention: the ban on made tea imports and green leaf exports, taxation, and the role of the Board and the ministries.

4.1 Import ban increases the burden on consumers

The ban on black and packed tea imports has been in place since the inception of the Tea Authority. While the ban on packed tea was imposed to protect the domestic blending and packing industry, the motivation behind the ban on black tea imports is not clear. Despite the ban, a substantial portion of the domestic tea market is supplied by imports from neighbouring tea producers – while there are no solid estimates on the amount of tea imports, industry and government representatives agree that it is between 30 and 35 per cent of domestic consumption. Industry representatives report that the imported tea is of secondary quality, in high demand among low income rural households. The imports continue because of this demand, the difficulty of monitoring trade in rural areas, and incentives on the supply side – importers' tax avoidance and exporters' avoidance of export taxes at home.

Opinions about the ‘tea problem’ and how to solve it are mixed. Ndunguru (2001: 7) has suggested that the Tea Board, the Tanzania Revenue Authority, and the Police should engage in a ‘war on illegal tea imports’. Others, however, have noted that what some see as a smuggling problem is really an excess taxation problem. Mdee (2001), for example, has noted that because tea importers avoid taxation they sell their tea much more cheaply than the locally produced tea.

The import ban has been seen as an infant industry protection measure: ‘The Tea Board of Tanzania has been waging a relentless war on this illegal practice and has made tremendous effort to protect the young industry, at least in these initial years before the industry grows and [is] in a position to compete with quality teas imported from outside’ (Mdee, 2001: 20). Yet the tea industry in Tanzania can hardly be considered young. Tea production was introduced in 1902 and blending and packing in 1970. If, as many report, the imported tea is of secondary quality and consumed mainly by low income rural households, the ‘tea problem’ is likely to be exacerbated as the quality of Tanzanian tea rises. Less domestically produced secondary quality tea will be available, boosting demand for lower quality imported tea. In a sense, this has happened already. Tanzania's average price discount over its major competitors has declined from 13 per cent during 1996–99 to 8 per cent during 2000–04, implying that there has been an increase in the quality of Tanzania's tea.

In addition to the demand side, i.e. poor consumers cannot afford high quality tea, there is a supply side argument. Consuming low quality tea locally while exporting high quality tea follows the logic that lower quality products are consumed close to the production point while quality rises the further away the consumer centre – for a full exposition of this issue see the ‘Shipping the Good Apples Out’ section in Silberberg (1978: 345).

East African tea producers have discussed the problem at the newly created East Africa Stakeholders Tea Committee, but no concrete solutions have been proposed. This is not surprising given the poor record of monitoring and enforcing bans and embargoes. As long as demand is strong and incentives remain on the supply side, attempts to end imports are not likely to succeed.

An alternative to the failed import ban would be to levy an import duty on tea of, say, 5 per cent. This policy would have several benefits. Some tax revenues would be generated, the quality of imported tea could be monitored, accurate statistics could be collected to improve policy making, legitimate jobs would be created for importers and traders, corruption would be reduced, and consumers could pay lower prices for tea. This policy action should be part of a comprehensive revision of the tax structure, which would ideally include a move to a uniform import duty across the board.

The export ban on green leaf tea should also be lifted. Article 27 of the 1999 Tea Regulations states that all green leaf tea produced in Tanzania shall be processed locally. While this restriction has no implications for the tea estates, it may prevent smallholders from selling their green leaf to processing facilities in neighbouring countries. Admittedly the prospects of exports of green leaf are slim at the moment. But, if this is the case, one may ask: why have a regulation in place which is irrelevant for all practical purposes?

4.2 Taxation is too complex and rates are too high

Despite frequent amendments to the tax code, it remains unnecessarily complicated. Currently there are as many as 44 taxes, levies, and licence fees. Moreover, tax rates are too high. The excess burden of taxation and the complicated nature of the tax code have been the subject of (and criticised by) various reports. Consider the following comments:

• ‘An excessive and complicated tax regime is a continuing growing concern of the tea industry’ (Ndunguru, 2001: 6).

• ‘New land rent tax levied on developed and underdeveloped land is too high and discourages tea expansion … raising fuel prices caused by excessive tax on the product. … Road toll tax on irrigation fuel is levied even though the fuel is not used for roads for vehicles’ (Brooke Bond Tanzania, Ltd).

• ‘The number of taxes [is] continuing to pose a growing concern to the tea industry. Taxes such as car benefit tax, payroll levy, education levy are eroding the income from the activities of the tea industry by raising tea production costs, thereby affecting the performance of the tea industry’ (Mdee, 2001: 22).

• ‘[S]till the traditional export crops are heavily taxed due to various fees …’ (Government of Tanzania, 1999: viii).

Taxes on the tea sector include a district produce tax of 5 per cent of the farmers' price, stamp duty of 1.2 per cent `free-on-board' (fob), withholding tax of 2 per cent fob, 3.5 per cent Tea Board and research fee, corporate tax, property tax, Value Added Tax, and a service levy of 0.3 per cent of VAT net turnover. The 2 per cent export tax was abolished in July 1998, and the agricultural land tax was reduced from Tsh600 (600 Tanzanian shillings) a hectare to Tsh200 a hectare – as of November 2001, the exchange rate was Tsh900 per $1. The duty on imported green and black tea is 25 per cent, down from 30 per cent before July 1998. The VAT on imported black tea is 20 per cent, down from 25 per cent before July 1998. The reduction in import duties and in the VAT on imported tea is irrelevant, however, because tea imports are prohibited.

The Government of Tanzania (1999) tax report – a comprehensive review of the tax structure of the export crop sectors – was unable to calculate the effective tax on the tea industry because it did not have a detailed representative sample. However, working backwards from tax payments, one can obtain an idea of the level of taxation. According to industry estimates, tea estates paid an estimated Tsh1.2 billion to the government in taxes and levies in 1999 (equivalent to $1.4 million at Tsh800 per $1). The estates produced about 20 000 tons of made tea during this year, fetching an average of $1.18 a kilogram in the Mombasa auction. An ex-factory price of about $1.00 a kilogram (approximately $20 million in gross revenue) implies an 8 per cent tax on gross revenue. Assuming a 25 per cent profit margin, the effective tax rate exceeds 30 per cent. A 10 per cent profit margin would imply an effective tax rate of 80 per cent. These high rates are consistent with findings by a recent Government of Tanzania report (2000) which estimated that the nominal protection coefficient of the tea sector was –55.2 in 1990–93 and –77.0 in 1994–99 (the negative sign indicates taxation).

The excess burden of taxation has numerous negative effects on the sector. The payroll levy and land tax, for example, discourage employment by effectively increasing the wage rate. High tax rates can lead to tax avoidance, tax evasion, corruption, and ultimately reduction in tax revenue (Laffer curve effect). And administering the taxes takes a substantial amount of staff time for producers as well as the government. Streamlining the tax code and reducing the tax burden should be a priority.

4.3 The Board and the ministries have too much discretionary power

Despite the restructuring of the tea sector, both the Tea Board and the Ministry of Agriculture still possess too much discretionary power. For example, Article 22 of the 1997 Tea Act indicates that the Tea Board may refuse licences on ‘any ground which may appear to it to be sufficient’. Moreover, Article 29 of the 1999 Tea Regulations specifies that in issues relating to tea quality for both domestic and export market the Board shall be the final arbitrator.

Wielding this excessive power, the Board has denied licences for imports of made tea and exports of green leaf as discussed earlier, a questionable policy on economic grounds and one that entirely ignores issues of consumer welfare. For example, Mdee (2001: 20) says about the import ban: ‘One of the functions of the Tea Board of Tanzania is to control the export and import of tea for the benefit of the farmers and the tea industry as a whole’. Yet the author is silent about the welfare implications to tea consumers, especially poor rural households. Regulations and trade policy decisions affecting the welfare of groups other than tea producers should not be the sole responsibility of the Tea Board but should be determined at a higher and broader level of policy making.

Penalties for violating Tea Board regulations are high – often as high as $2 000, which is ten times Tanzania's per capita GDP, equivalent to a fine of $300 000 in the United States. Despite such stiff penalties a third of Tanzania's domestic tea demand is supplied by ‘prohibited’ imports, implying that tea smugglers are not caught (the ban is ineffective) or are caught and not fined (the ban spawns corruption). In addition, violations of rules and regulations by the government are frequent. For example, the 1997 Tea Act (p 13) stipulates that the Tea Association has the right to nominate two members to its board: ‘The Board of Directors shall consist of … (d) two other members representing the interests of licences nominated by the Tea Association of Tanzania amongst farmers or manufacturers’. However, as the government review of the agriculture sector noted: ‘The President did not respect this legal provision in a reshuffle of the [Tea and Sisal] Crop Board members in June 1999. Because of that reshuffle most of the members of the Coffee, Cotton, and Cashewnut Associations, who were “a titre personnel” [sic] member of the crop Boards, lost their membership of the Board as well’ (Government of Tanzania, 2000: 124).

5. Conclusion

A number of important conclusions emerge from the preceding review. First, reforms of the tea sector started much earlier than reforms in other export crop sectors. Second, it appears that there was no external or ‘push’ factor for the tea reforms. They were undertaken, and hence owned, by the government and the industry. Third, the reforms have been, by and large, successful. This is in stark contrast to reforms in the other export crop sectors where reforms did not start until the mid-1990s, where there was considerable donor assistance (either technical or financial), and where the reforms have not been as successful (see, for example, Baffes, 2004b, 2005; Mitchell, 2004, for review of Tanzania's cotton, coffee, and cashew sectors).

The success of reforms can be attributed to a number of factors. For one, both the two nationalised estates and the smallholder sector reached a state of collapse. Therefore reforms appeared not only to be the only feasible alternative but there was also no opposition to reforms since there were no potential losers. On the other hand, the fact that the estates which were under private management and ownership were very successful indicated that the poor performance of the nationalised estates and smallholder sector reflected bad management and poor policy choices rather than external factors. Stated simply, there were no excuses not to reform.

Despite the successful outcomes, the reform agenda is by no means complete. Reducing taxation and streamlining the tax code will certainly induce further supply response as more resources will remain within the sector. Reforming trade policy is also essential. In addition to the benefits to be realised by poor rural households, reducing the tea industry's trade barriers will increase regional cooperation and also signal good-will, thus contributing (albeit in a very minor way) to a successful Doha Development Agenda outcome of the multilateral trade negotiations. Finally, the power of the Tea Board and the relevant ministries should be limited to activities such as collecting and disseminating market information and statistics and enforcing regulations that contribute to a friendly investment climate.

The views expressed here are those of the author and should not be attributed to the World Bank. I would like to thank Takamasa Akiyama, Betty Dow, Donald Mitchell, Hans Timmer, and Nanae Yabuki for comments and suggestions on earlier drafts. I am grateful to Ria Ketting, George Kyejo, SH Mijinga, Bruno J Ndunguru, Hadija Shakombo, RJ Surrey, and staff at the East Usambaras Tea Company for the valuable information obtained during interviews. I would also like to thank the World Bank country office staff in Dar es Salaam, especially Ladisy Chengula and Gloria Sindano.

Notes

An earlier version of this paper (which also contains an extensive discussion of the world tea market) has been published in the World Bank's Africa Working Paper series (Baffes, 2004a).