Malawi's Tobacco Sector
   Standing on One Strong Leg
     is Better Than on None




            Steven Jaffee




              June 2003

Malawi' s Tobacco Sector: Standing on One Strong Leg is Better Than on None
Africa Region Working Paper Series No. 55
June 2003

Abstract

T    obacco plays a central role in Malawi's               economy. Initial signs of this were evident in the late
     economy, accounting for 60% of its exports,           1990s.
13% of its GDP, and 23% of its total tax base. One in
five Malawian households rely substantially upon           Yet, this engine for growth and diversification has
income from tobacco production or employment.              since stalled. The earlier flow of cash income into rural
Malawi, a small, land-locked country, is the world's       areas has slowed to a trickle in recent years as a result
leading exporter of burley tobacco leaf, having            of sharp tobacco price declines (related to short-term
recently overtaken the exports of the United States.       international supply and demand conditions), reduced
Despite international efforts to reduce cigarette use      farm productivity and high and rising transaction and
because of its adverse health effects, world demand        logistics costs in the tobacco marketing chain.
for Malawi's tobacco should remain robust for many         Institutional limitations and conflicting interest within
years as a result of the expected continued growth in      the industry have inhibited the development of an
consumption in many low and middle income                  appropriate policy and operational response to the
countries   and   as   a   result of   the   favorable     productivity    problems    and   loss   of  profitability
characteristics and competitive cost of Malawi's           experienced by many Malawian growers. This paper
tobacco.                                                   highlights the challenges facing the industry and
                                                           Malawian policy-makers and outlines options for
Still, Malawi faces substantial challenges in drawing      moving forward in addressing those challenges.
sustained    benefits  from    its  relatively  strong     Malawi's capacity to achieve its ambitions for
international market position. The liberalization of       economic diversification depend, in part, on its ability
the sector during the early to mid-1990s resulted in a     to overcome the current challenges facing its dominant
massive entry of smallholder farmers into tobacco          industry. It is a more profitable tobacco industry--
production. While the income from production was           rather than wishful thinking--which can serve as an
formerly concentrated among a limited number of            effective spring board for move diversified patterns of
commercial estate owners, such income was now              investment, employment and trade in Malawi.
spread among some 350,000 households, with
powerful multiplier effects in rural areas. There was      This paper was prepared as a contribution to the
an expectation that these developments would create        Malawi Country Economic Memo randum and to the
an institutional and financial platform for the            Malawi Diagnostic Trade Study under the Integrated
diversification of the rural and broader Malawian          Framework.




Author

Steven M. Jaffee, Sr. Economist, PRMTR
Sjaffee@worldbank.org



The Africa Region Working Paper Series expedites dissemination of applied research and policy studies with
potential for improving economic performance and social conditions in Sub-Saharan Africa. The Series publishes
papers at preliminary stages to stimulate timely discussion within the Region and among client countries, donors,
and the policy research community. The editorial board for the Series consists of representatives from professional
families appointed by the Region's Sector Directors. For additional information, please contact Paula White,
managing     editor   of  the   series,  (81131),   Email: pwhite2@worldbank.org        or  visit  the  Web    site:
http://www.worldbank.org/afr/wps/index.htm.


The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s), they
do not necessarily represent the views of the World Bank Group, its Executive Directors, or the countries
they represent and should not be attributed to them.

                                                 Table of Contents


Introduction .................................................................................................................................1

  Background ................................................................................................................................. 1
  Purpose and Content of the Paper............................................................................................... 4
  The World Bank, Tobacco, and Economic Diversification........................................................ 5

Trends in International Supply and Demand for Tobacco.............................................6

  Production Trends....................................................................................................................... 6
  Demand Trends........................................................................................................................... 9
  Trade Trends ............................................................................................................................. 10

Major Recent Trends in Malawi's Tobacco Subsector .................................................12

  Production Trends..................................................................................................................... 12
  Continued Structural Change .................................................................................................... 14
  Trends in Prices, Quality, and Productivity.............................................................................. 17
  Declining Profitability............................................................................................................... 20
  Production Cost Economies...................................................................................................... 25
  Exports and Gross Revenues .................................................................................................... 27

Major Issues and Policy Options.........................................................................................29

  Diversify From a Strong Base................................................................................................... 29
  Core Agenda Summarized ........................................................................................................ 33
  Allowing Direct Exports by Growers: Is the Grass Greener on the Other Side? ..................... 35
  Reducing Marketing, Logistical and Other Transaction Costs................................................. 39
  Raising Productivity.................................................................................................................. 50

Conclusions...............................................................................................................................51

References.................................................................................................................................52


                                                       List of Tables

Table 1: Major Tobacco Producing Countries : Comparative Structural Features.................................. 9
Table 2: Exports of Tobacco Leaf (000 MT)..................................................................................... 10
Table 3: Malawi Tobacco Production Trends (OOO Tons)................................................................ 13
Table 4: Structural Transformation of Malawi Burley Tobacco Production (Author's Estimates)........ 15
Table 5: Net Returns for (NASFAM) Smallholder Burley Growers.................................................... 21
Table 6: Exchange Rate Movements Relevant to Tobacco Earnings ................................................... 22
Table 7: Comparative Production Costs for Burley Tobacco.............................................................. 26
Table 8: Leading Tobacco Leaf Exporters: Shares of National Exports .............................................. 29
Table 9: From Auction Floor toOverseas Market (2001): Cost Build-up of Burley
    Tobacco (US Cents/kg.)........................................................................................................ 37
Table 10: Intermediate Buyers: Performance on Burley Tobacco Sales, 1994 � 2001 .......................... 45
Table 11: Tobacco Marketing Arrangements for Leading World Producers........................................ 48

                                           List of Figures

Figure 1:Average World Price for Exported Tobacco Leaf ($Per Ton) ....................................... 11
Figure 2: Recent Trends in Average Tobacco Auction Prices in Malawi.................................... 18
Figure 3: Producer Prices for Burley Tobacco ............................................................................. 18
Figure 4: Comparative Tobacco Yields (KG/HA)........................................................................ 20
Figure 5: Commercial Lending for Agricultural in Real Terms ................................................... 25
Figure 6:Smallholder Burley Production Costs (US$ per Hectare).............................................. 26
Figure 7: Malawi Tobacco Leaf Exports ...................................................................................... 27
Figure 8: Declining Revenue Indicators: Burley + Flue Cured .................................................... 28
Figure 9: Auction Level Payments as % of Tobacco Export Earnings........................................ 36

                                            List of Boxes

Box 1: Malawi without Burley Liberalization................................................................................... 30
Box 2: Exit and Voice in Malawi's Tobacco Subsector............................................................... 31
Box 3: SWOT Analysis for Malawi's Tobacco Subsector........................................................... 33


Annex: 1: Quality Results of TAMA Tobacco Classification System Results
    for Burley Tobacco, 1992 to 2001............................................................................................. 54

                                           INTRODUCTION


Background


        The commercial cultivation of tobacco has a long history in Malawi, dating back to the
1890s.1 By the 1920s, tobacco assumed a prominent place in the economy of what was then
Nyasaland Protectorate. Its expansion and contractions remained a major feature of the colonial
economy in Nyasaland. In the period following Malawi's independence in 1964, tobacco took on
an even more central role in the political economy of the country, as large segments of the
nation's political leadership and economic elite invested in the crop, backed by supportive
policies, regulations and public resources. With the rapid expansion of the industry in the 1970s
and 1980s, tobacco became the country's primary source of wealth, political patronage, (non-
governmental) employment and foreign exchange earnings. The internal and cross-border
migration of people to fulfill the labor requirements for the expanding industry even had a
profound effect on national demographic patterns.2

        Previous research has examined the performance and structural changes of Malawi's
tobacco industry during the 1970s and 1980s as well as captured some of the early trends which
emerged from the liberalization of the subsector in the early to mid-1990s. During the colonial
period and from independence through to the late 1970s, participation in the lucrative cultivation
of tobacco was restricted to an elite cadre of growers who owned or leased estates. These farmers
were permitted to sell their tobacco directly to international buyers at officially recognized local
auctions. Smallholder farmers were restricted in the varieties of tobacco which they could
cultivate and were restricted to whom they could sell. For the most part, smallholder farmers
were required to sell to government agencies3 and received prices well below prevailing market
prices.

        Most of the opportunities for smallholders to participate in the sector were as tenants or
other types of laborers on the estates. With few remunerative opportunities available to this
population, such labor contracts became widespread. The Special Crops Act (1964) maintained
earlier restrictions on smallholder farmers, preventing them, with few exceptions, from the
cultivation of the higher value burley and flue-cured varieties of tobacco. The system of
production controls remained firmly in place through the 1970s and 1980s. This, together with a
system of production quotas--ostensibly designed to control the overall size of the tobacco crop
to approximate the committed demand of major international buyers--served as a primary means
of allocating opportunities and distributing income and wealth in the country.

        Apparently in recognition that the elite-centered growth model was fostering some
dissent while also failing to foster the wider development of the economy, the political


1See Wilshaw's A Century of Growth: Malawi's Tobacco Industry, 1893-1993.
2Kydd and Christiansen (1982); Jaffee et al. (1991)
3The Native Tobacco Board during the colonial period and ADMARC since independence.



                                                      1

leadership accommodated and in fact encouraged the emergence of an additional wave of
Kulack-type entrepreneurs during the 1980s from the midst of the small business and small
farmer communities. These families (or extended families) were enabled to acquire leases on
formerly customary land and in establishing `estates' (typically 10 to 20 hectares) these
households were permitted to grow burley or flue-cured tobacco and permitted to sell their
tobacco directly through the auctions rather than through a government agency.

         Some 30,000 of these mini-estates were formed during the 1980s, with this development
accounting for much of the growth that was recorded in the industry during that decade.4
Nevertheless, the bulk of Malawi's rural population and farmers were still prevented from
directly participating in the expanding subsector other than as laborers, or, increasingly, as illegal
growers who sold their crop cheaply over the farm-gate to estate owners and quota holders.

         In 1993 a new government was elected in Malawi, following nearly three decades of one-
party rule. The vision of the new government was one of more broadly based growth, with
increased attention to be given to poverty alleviation. This orientation, together with strong
pressures made by several development donors, resulted in steps to liberalize the tobacco
subsector. The new government moved to amend the Special Crops Act, allocate (expanding)
tobacco sales quotas to groups of smallholder farmers (who were organized into `clubs'), and
introduced a program of intermediate buyers in order to facilitate the logistics of bringing the
smallholder crop to the auctions and to increase competition in the purchasing of the smallholder
tobacco crop.

         Vested interests within the estate sub-sector lobbied (unsuccessfully) against the tobacco-
related reforms, fearing both future competition in the tobacco market as well as the potential
loss of skilled laborers who might chose to take up tobacco cultivation on their own small plots.
Arguments were made that smallholders would undermine the quality reputation of the Malawi
tobacco crop. However, while a growing number of estates were experiencing financial and/or
managerial problems (even in the face of favorable tobacco prices during the early 1990s), most
stakeholders anticipated a gradual, long-term structural change within the industry and it was
generally assumed that well-managed estates would continue to have a long-term future in the
industry.5

         Research carried out in the mid-to-late1990s emphasized the relatively rapid and large
initial response by smallholders to the policy reforms (with up to 200,000 smallholders taking up
tobacco cultivation by 1996) and the significant effects which tobacco-derived incomes were
beginning to have on household expenditures and savings and on rural communities within the
main tobacco-growing areas.6 Related work demonstrated the relatively high profitability of
tobacco cultivation in Malawi, compared with that for most other cash and food crops.7 These
and other8 studies pointed to the divergent patterns of productivity and profitability within the

4This process in analyzed in depth in Mkandawire et al. (1990).
5As Banda et al. (1998) noted, it was assumed that there would be "gradual transformation without cataclysmic
change."
6Zeller (1997) and Jaffee (1997)
7Keyser (1997)
8Gossage et al. (1997)



                                                         2

estate sub-sector and suggested that significant numbers of estates might be unable to adjust and
remain viable should tobacco prices decline significantly. At the same time, questions were
raised about the sustainability of the smallholder tobacco boom and whether the underlying
institutional arrangements governing production support services and tobacco marketing were
adequate or even fully appropriate given the evolving structure of the industry.9

        During the last few years, many of the concerns about estate viability and smallholder
production sustainability have come to fruition. However, through much of this period there has
been a state of policy paralysis characterized by an on-going struggle to maintain, undermine, or
otherwise re-direct the earlier policy reforms within the industry. Many stakeholders have
seemingly concluded that the best days of Malawi's tobacco industry are behind it and that the
current game is a zero-sum one in which (reduced) incomes and opportunities must be fought
over. All parties �farmers, intermediate buyers, tobacco buyers, the auctioning company,
financial institutions, the government, etc.--have been subjected to accusations about how their
actions or inactions have weakened the industry. It has not been a very productive dialogue and
this has delayed concerted action to address the major underlying problems within the industry.

        The problem has not simply been one of conflicting interests, although these and an
imbalanced power structure within the industry have certainly been factors in the apparent
paralysis. Another important factor has been the inadequacy, non-transparency, and asymmetric
distribution of data and information regarding the performance of the industry and of different
stakeholders therein. Most participating stakeholders have only partial information, few local
players have comparative competitor country performance data, and significant parts of the
official statistics for the industry--both on physical quantities and on an array of
financial/economic matters may not provide an accurate picture of the prevailing situation.
Performance monitoring arrangements and the underlying registration system for tobacco
producers have simply not kept pace with the demands of a rapidly changing industry. Hence, for
the most part, major stakeholders have not been working from the same sheet of music and there
has been an inability to reach consensus on what is happening let alone on what needs to be done
to improve the industry's competitiveness and viability.

        Only in the past two years--years of particular difficulty within the industry--has there
been a change in the tone and content of subsector consultations and policy dialogue, with some
improvement in the mutual understanding among different stakeholders' constraints. As this
paper will illustrate, the Malawi tobacco subsector lies at a critical juncture, facing major
problems which threaten its sustainability and which pose a number of serious policy questions.
Addressing these issues effectively will require the intensification of the type of constructive
dialogue which has been apparent more recently.

        The importance of tobacco to Malawi's economy cannot be exaggerated and there are
relatively few examples elsewhere of a similar level of dependence on one sub-sector in a
country's exports earnings, employment, and income flows. Tobacco accounts for some 60% of
Malawi's merchandise exports, some 23% of its total tax base, and as much as 13% of its GDP.
Many of Malawi's cities have been built by tobacco wealth while most of the few signs of


9See, especially, Banda et al. (1998)



                                                 3

`prosperity' which one sees in rural Malawi (i.e. tin roofs, bicycles, radios) have like-wise been
substantially generated by tobacco-related incomes.


Purpose and Content of the Paper


         This paper seeks to provide a summary of major trends and developments within
Malawi's tobacco subsector since the mid-1990s and to elucidate the policy and institutional
options seemingly available to address a range of constraints/problems currently facing the
subsector. The paper is not based on extensive fieldwork, but rather on a limited range of
interviews with major stakeholders and organizations, available data and position papers, and
data and information on international and competitor country trends, derived from a variety of
sources. The paper seeks to further stimulate and inform the on-going policy debates related to
Malawi tobacco and its role in the wider economy there.

The leading questions addressed in this paper are the following:

     � What are major trends in the international demand and supply of tobacco (and tobacco
         products) and what are the near and medium term prospects?

     � How is Malawi positioned in that international market? Are concerns about this being a
         `sunset' industry for Malawi well founded?
     � What have been the patterns of tobacco-related structural change since the mid-1990s and
         what factors have driven such changes?10
     � What has been the performance of the subsector during this period in terms of
         productivity, earnings, product quality, trade, etc.? How does this performance compare
         with that of other major tobacco-producing countries?

     � What factors have influenced the underlying performance of the subsector? How
         effective have been the institutional arrangements which were put in place during the
         mid-1990s to complement the liberalization of the subsector?
     � What are the main institutional and other constraints currently facing Malawi's tobacco
         growers, especially smallholder farmers?
     � What are the policy and other options for addressing these constraints and what are the
         possible benefits, costs and risks of adopting selected approaches?
     � What measures are recommended for short and medium term action?

The paper is organized as follows. The next section provides an overview of trends in
international supply and demand for tobacco, drawing upon FAO, USDA, and other sources.
This is followed by a review and analysis of major trends occurring within Malawi's tobacco
subsector since the mid-1990s. This covers trends in production, exports, pricing, profitability,

10Unfortunately, we are not in position to examine some of the medium-term impacts which tobacco production and
sales have had on smallholder households and communities. Some work on this was undertaken in the mid-1990s,
yet there have been little or no household/community studies done in recent years to expressly examine the impacts
which the spread of tobacco cultivation (to nearly 20% of rural households) has had.



                                                        4

productivity, etc. The subsequent section then examines a range of policy questions and options
that have been recently or are currently facing the subsector. Some recommendations are made.
A short conclusions section follows.


The World Bank, Tobacco, and Economic Diversification


         Since 1991, the World Bank has had a formal policy that it will not lend directly for,
invest in, or guarantee investments or loans for tobacco production, processing, or marketing.
Bank loans also cannot be used to finance imports of tobacco or tobacco products, tobacco
processing machinery or other related equipment or services. Bank activities in the health sector
discourage the use of tobacco products. The Bank has been an active participant in international
efforts to reduce demand for tobacco products, especially in client countries and to induce
countries to adopt policies which curb tobacco products use and internalize the externalities
associated with their use.11

         This general policy poses a major challenge to the Bank's support to Malawi's economic
development. Both the original (1991) and the recently updated (1999) "operational policy" on
this issue notes that exceptions may be made for countries that are heavily dependent upon
tobacco as a source of income and foreign exchange (i.e. more than 10% of exports). Only two
client countries have such a high dependence on tobacco for foreign earnings--namely
Zimbabwe (20%) and Malawi (60%).

         Given the scale and spread of Malawi's industry any support for infrastructure
development, financial systems development and even entrepreneurship development provides
some backing to the tobacco sub-sector there. The same applies to support provided for
agricultural research, extension, and policy planning. But, the Bank has gone further, recognizing
in the late 1980s and early 1990s that promoting broad based growth and poverty reduction in
Malawi was simply not possible without fostering more remunerative participation by the poor in
this dominant industry.12

         The Bank and other donor agencies have and will continue to provide support to facilitate
the diversification of the rural and national economy in Malawi. However, given the tiny size of
the domestic market, the country's land-locked position, and its human and natural resource
base, this diversification will be a long-term process. Its success almost certainly depends upon


11The Bank's work on tobacco control is in close partnership with the WHO, the UN Foundation, the US CDC,
PAHO, and others. Given the Bank's comparative advantage in economics and policy dialogue, its efforts have
focused on the economics of tobacco control, including taxation and the economic and social impact of control
measures. See www.worldbank.org/tobacco
12In its 1997 Growth Prospects Paper, the Bank recognized the importance of the tobacco liberalization process for
fostering broad based growth in the near term via the direct incomes/employment and the multipliers expected from
higher purchasing power by large numbers of smallholder households. Other, less certain, pillars for medium term
growth were envisioned to include productivity gains in food crop agriculture, and possible investment in tourism
and labor-intensive manufacturing. Since that time, Malawi's manufacturing sector has contracted and the country
has had little success competing for regional and international tourists within the southern Africa tourism market.
There have been some productivity gains in food crop production by smallholders, yet their magnitude is uncertain
and some of the gains have been dependent upon the (donor-financed) free distribution of fertilizer and seeds.



                                                           5

the maintenance of a competitive and profitable tobacco subsector from which savings can be
derived and invested in other economic activities. Malawi must therefore continue to exploit its
comparative and competitive advantages in tobacco at least over the medium term. The country
simply has no (realistic) choice.13


                           TRENDS IN INTERNATIONAL SUPPLY
                                 AND DEMAND FOR TOBACCO

Production Trends


         Over the period from 1980 to 2001 world production of tobacco leaf increased from 5.26
million tons to 6.97 million tons, or at 1.02% per annum. Most of this growth occurred in the
1980s and the early 1990s. There was a slowdown thereafter, partly because of the build-up of
excessive unsold stocks in several countries, including China, India, and Turkey. World
production grew at 1.95% per annum during the 1980s, yet declined by 0.69% per annum over
the 1990 to 2001 period.14

         The expansion in production over this two-decade period was entirely accounted for by
developing countries. Tobacco production in developed countries--including the United States,
European Union and Japan--declined from 1.99 million tons to 1.29 million tons over this
period. Production in these countries declined by 2.0% per annum during the 1980s and by 3.1%
per annum in the period since 1990. In sharp contrast, tobacco production in developing
countries increased from 3.26 million tons in 1980 to 5.68 million tons in 2001. Their share of
global production therefore rose from 62% to 81% over this period.                         Developing country
production grew by 3.8 % per annum during the 1980s, while recording no growth during the
1990s. The latter figure clouds a trend in which production in China leveled off or even declined
in some years while output continued to expand in such countries as Brazil, India, Zimbabwe
(until the past two years) and Malawi.

Major factors underpinning these divergent trends have included:

    � Shifts in the locus of demand from developed to developing countries (see below)
    � Declining political and taxpayer support for tobacco production in developed countries,
         including the United States and the European Union;
    � The cost competitiveness of developing countries and the relatively high profitability of
         tobacco compared with that of many other crops in these countries;
    � Direct technical and financial support for tobacco cultivation in several developing
         countries; and

13Unless, perhaps, the international community is prepared to `buy-out' Malawi's tobacco growers in a manner
similar to the approach being used in the United States to wean tobacco growers (and regions) of their dependence
on tobacco incomes and employment. Even if the international community were prepared to do this there still
remains a question regarding the industries or subsectors in which Malawi could achieve a significant scale and
position of international competitiveness within a medium term time frame.
14Based on FAO Production Statistics.



                                                          6

    � Investment by international companies in the promotion of tobacco production and in
             cigarette manufacturing in selected developing countries.

    The expansion of tobacco production in developing countries has occurred through a
combination of area expansion (especially in China and Malawi) and productivity gains. Over
the 1980 to 2001 period, the tobacco area harvested in developing countries increased by 1.46%
per annum while yields grew at 0.87% per annum. During the more recent 1990 � 2001 period,
developing country plantings actually contracted by 0.68% per annum, although production was
maintained as a result of a 0.69% per annum increase in yields. These trends are summarized in
the graph below.


                      Developing Countries: Tobacco Area and Yield Trends


             5,000,000                                                                                       1,700
             4,500,000                                                                                       1,600
             4,000,000                                                                                       1,500
             3,500,000                                                                                       1,400
             3,000,000                                                                                       1,300
     Hectares2,500,000                                                                                       1,200 (Kg/Ha)

             2,000,000                                                                                       1,100
             1,500,000                                                                                       1,000
                                                                                                                            Area Harvested
                                                                                                                            Yield
                    1980198119821983198419851986198719881989199019911992199319941995199619971998199920002001


             Although many countries produce some tobacco leaf, world production is highly
concentrated. China alone accounts for 38% of world production while three other countries--
India, Brazil, and the United States--in combination account for another 25% of world
production. Seven other countries--Turkey, Zimbabwe, Indonesia, Italy, Greece, Malawi, and
Argentina--also produce more than 100,000 tons per year of tobacco leaf and have a combined
global share of 15%. Hence, eleven countries account for nearly 80% of world tobacco
production. See the graph below.

             Many different varieties of tobacco leaf are produced and the concentration of production
is even more evident for certain tobacco types. The most widely grown type of tobacco is Flue-
cured Virginia, a flavorful tobacco which is cured in small ovens using wood or coal for fuel.
Just under 60% of tobacco production is of varieties of this type. China accounts for more than
half of the world's production of flue-cured tobacco, with other major producers being the
United States, Brazil, Zimbabwe, and India. Burley tobacco, an air-cured variety, accounts for
just under 15% of global tobacco production. The largest producers of burley tobacco are the
United States and Malawi, followed by China and Brazil. Oriental tobacco, with strong aromatic
qualities, accounts for about 10% of total tobacco production. Turkey is the world leader (40% of
the total), followed by Greece. Most of the balance of world production is accounted for by a
number of other sun and air-cured varieties of tobacco which are generally used in hand-rolled
cigarettes. These are widely produced in Asia, especially in India (for bidis) and in Indonesia
(kretek).


                                                                                  7

                  Leading Tobacco Producers 2001 (000 MT)
                                                                                 China
                                                                                 India
                                                                                 Brazil
                              1524
                                                                                 United States
                          101                          2661                      Turkey
                          120                                                    Zimbabwe

                           130                                                   Indonesia
                           132
                           147                                                   Italy
                           175                                                   Greece
                             260
                                450                                              Malawi
                                      569         702                            Argentina
                                                                                 Others


         Worldwide, probably between eleven and twelve million farmers cultivate tobacco, with
about two-thirds of these being in China. While there are economies of scale in the processing
and logistical management of tobacco leaf and in the manufacturing of cigarettes, there are few
economies of scale in the actual cultivation of the crop. Tobacco requires very careful crop
husbandry and the tobacco leaves are harvested by hand, on several occasions as the plant
growth progresses. Only the land preparation stage is amenable to mechanization. Largely for
these reasons, tobacco is typically cultivated on small family farms. On most such farms, the
typical plantings of tobacco are a few hectares. In such countries as China, Turkey and (recently
in) Malawi, the typical tobacco planting is 0.5 ha. or less. The structural patterns in seven of the
leading tobacco producing countries are summarized in Table 1.

         Conventional cigarettes are by far the largest form of tobacco use in the world. As with
tobacco leaf, there has also occurred a significant shift in the locus of their production. Once
manufactured predominantly in developing countries, since the mid-1980s there has been rapidly
growing production in developing countries, both as a result of significant developments within
China and the relocation by major multinationals to better position themselves in growing
consumer markets.

         Among cigarettes, there is an emerging trend toward greater consumption of American
blend types which are produced with a mixture of Virginia, Burley, and Oriental tobaccos.
According to industry sources, world cigarette production was essentially the same in 2000 (at
5.5 trillion units) as it was in 1990 (5.42 trillion units). However, over this period, the production
of American blend cigarettes increased by 12%, while declines were recorded in the production
of purely Virginia (-4%), oriental     (-16%), and dark/Kretek-type (-9%) cigarettes. As a result,
the share of American Blend cigarettes in the total world production of cigarettes increased from
33% to 43% over this period.




                                                    8

           Table 1: Major Tobacco Producing Countries : Comparative Structural Features

                          China           Brazil     USA    Zimbabwe       India       Turkey        Malawi
      Production          2600            590        480    210            660         260           125
      (000 MT)
      Planted Area        1600            330        195    80             435         280           140
      (000 Ha)
      # of Growers        8000            135        90     18             850         576           375
      (000)
      Average Size of     0.4             16.8       66     300+           2.5         4.9           1.0
      `Tobacco' Farm
      (Ha)
      Average             0.2             2.6        4.2    40             1.3         0.5           0.2
      Tobacco
      Planting (Ha)
      Tobacco             90% FCV         75%        2/3    95% FCV        75%         95% Ori       92% Bur
      Varieties*          9% Bur          FCV        FCV                   Non-
                                          16% Bur    1/3                   cigarette
                                                     Bur                   varieties
                                                                           FCV +
                                                                           Bur 5%
      * Flue-Cured Virginia (FCV); Burley (BUR); Oriental (ORI)
      Sources: USDA Attache Reports; Kasnakoglu and Cakmak (2000)


Demand Trends


         World demand for tobacco products continues to rise. Divergent patterns are again
apparent with per capita and even absolute declines in consumption in developed countries and
growth in developing countries. The growth in the latter countries has largely been fueled by
population and income growth, together with the marketing efforts of cigarette manufacturers.
Other factors influencing demand have been prices, habits, taxation, information on the health
risks of tobacco use, and restrictions on smoking in certain countries. Developing countries now
account for 70% of world consumption with China alone accounting for 44%.

         According to Mergos (2001) world demand for tobacco products is expected to increase
by 2% per annum between 1998 and 2005 and accelerate to a 2.3% per annum growth rate over
the subsequent five years.15 The divergent trends between developed and developing countries
are expected to continue. While tobacco consumption is expected to increase by 3.2% per annum
this decade in developing countries, demand in developed countries is expected to fall by 0.7%
per annum between 1998 and 2005 and by 0.3% between 2005 and 2010. The Asian region is
expected to account for the bulk of the growth in world demand during the coming decade.




15Mergos' projections should probably be regarded as the upper limit on expected growth in world demand. His
projections appear to be extrapolated from trends in demand, prices, population growth, and income growth over the
1980 to 2000 period. In the latter part of the 1990s there was some slowdown in the growth rate of developing
country tobacco demand. While some of this slowly was almost certainly due to the economic crisis in parts of
Southeast Asia, it is possible that the slowdown also reflects some degree of effectiveness in country campaigns to
increase awareness about the health risks of tobacco use.



                                                           9

         As in the recent past, this growth in demand will be fueled significantly by population
and income growth. These trends are expected to overwhelm the demand-depressing effects of
increased taxes or non-financial programs (i.e. advertising bans) that may be implemented in
developing countries, promoted by the WHO Framework Convention for Tobacco Control.16                             17

Longer term, the number of smokers is expected to increase to 1.6 billion (from 1.1 billion
today) by 2025.18


Trade Trends


         During the 1980s and early 90s, some 22 to 24% of world tobacco leaf production was
traded internationally. This share has increased somewhat in recent years, being just under 29%
in 1998-1999. While in the 1980s, the volume and value of world tobacco leaf exports increased
by 0.22% and 0.64% per annum, respectively, during the 1990s, annual growth in the volume of
world leaf exports was just over 3% while the value of such trade increased by 2.5% per annum.
As with production and demand, it has been developing countries which have achieved increased
shares of world trade with the most prominent gains achieved by Brazil, Zimbabwe and Malawi,
at the expense of traditional suppliers, including the United States, Greece, and Italy. These
trends are summarized in the table below.


                                  Table 2: Exports of Tobacco Leaf (000 MT)

     Countries                           1980-82                   1990-92                     1997-99
     Developed                             596                        731                         745
     Developing                            814                       1011                        1263

     USA                                   265                        235                         212
     Italy                                  73                        128                         96
     Greece                                 64                        119                         68

     Brazil                                153                        226                         321
     Zimbabwe                              103                        155                         174
     Turkey                                107                        102                         149
     Malawi                                 49                         97                         113
     Source: FAO Trade Statistics

         During the 1980s, world tobacco (average export) prices remained within a relatively
narrow range of $2800 to $3100/ton.19 During the 1990s there was somewhat more volatility in
world tobacco prices, yet still such prices were far more stable than was the case for many other
agricultural commodities, especially such important African export crops as coffee, cocoa, and


16The Framework will be an international legal instrument intended to limit the harm to health caused by tobacco
products. It is expected to address tobacco advertising and promotion, regulation, smuggling, excise tax levels,
treatment of tobacco dependence, and smoke-free areas.
17Many countries have been implementing tobacco control measures for decades, with varying levels of
commitment and impact.
18Mergos (2001)
19The only outlier years were 1982 and 1983 when the average values were $3235 and $3126/ton.



                                                          10

cotton. To illustrate this, over the 1990 to 2000 period, the coefficient of variation in
international tobacco prices was 0.07, compared with 0.35, 0.18, and 0.19 for Arabica coffee,
cocoa, and cotton, respectively.

        Nevertheless, the graph below does indicate that in the period since 1997 there has been a
major (i.e. 20%) decline in the average world price for exported tobacco leaf. During this period
there was an overall imbalance in global supply and demand, partly due to the economic
recession faced in many parts of the world (most significantly in Asia) and partly due to
expanded production and, especially, the build up of large quantities of tobacco leaf stocks, most
prominently in China, India, and Turkey. Uncertainty surrounding the legal proceedings against
cigarette manufacturers and the financial restitutions which would come out of such proceedings
may also have put downward pressures on international tobacco leaf prices.20

        With several countries destroying unsold stocks and cutting back on production (i.e.
India's moratorium on flue-cured production for the 2001/2002 season) and with the disruption
in production in Zimbabwe, the near-term situation appears to be more favorable than in the
recent past with regards to the overall balance of international supply and demand. Some modest
improvement in prices is therefore expected in the short-term.

                    Figure 1:Average World Price for Exported Tobacco Leaf ($Per Ton)


     3750
     3500
     3250
     3000
     2750
     2500
               1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001


        Over the medium term, there is some uncertainly about the likely direction of
international prices, although these are more likely to be stagnant or somewhat declining rather
than experience an increase. Unless there is a massive unexpected decline in the production of
one or more major producers, it is most unlikely that world prices will revisit the levels achieved
in the early 1990s and again in 1996-97. On the supply side, there is an expectation that both US
and EU tobacco production will contract (as price supports are cut back and possibly removed)
as will their supplies into the international market. Considerable uncertainty relates to the future
prospects for Zimbabwe as a tobacco producer/exporter as it is not clear whether the current
instabilities there represent a temporary blip for the tobacco industry or will result in a long-term
reduction in the capacities of that industry.




20Even though tobacco leaf itself accounts for only a small proportion of the cost and price of cigarettes.



                                                        11

        Of the other major producers, the prospects are probably brightest for Brazil and China.
Brazil produces a different style of tobacco than Malawi, yet any large future expansion in Brazil
could impinge on Malawi's market position. China is a very large producer but not a regular
exporter of tobacco. While the current thinking is that China will increase its tobacco imports (of
particular styles and varieties) there remains a risk that China will, periodically, put large
quantities of current production or accumulated stocks into the market, potentially depressing
prices for others.

        According to industry sources, Malawi is relatively well positioned in the international
tobacco market. The country is the second largest producer and exporter of burley tobacco,
behind the United States. It does not directly compete with the United States as that country
produces a relatively high-quality and very high cost aromatic burley tobacco. Malawi produces
a burley tobacco which has excellent `filler' properties. It is amongst the cleanest tobaccos in the
world because there is very little agro-chemical use in Malawi. Malawi's tobacco also has
relatively low nicotine content, this being a valuable attribute for a `filler-type' tobacco.

        Malawi also does not directly compete with Brazil or with India as the former generally
produces a high nicotine, flavored tobacco, while India produces very little burley tobacco and
one which is generally of low quality. There are a number of countries which do produce styles
of tobacco leaf similar to that of Malawi--including Thailand and Mexico--yet these countries
generally produce for their domestic markets and have only limited exports. Zimbabwe is a
major player in the flue-cured tobacco market and not in the burley market. Some up and coming
producers, including Zambia and Mozambique, produce a similar style burley crop as Malawi,
yet their production is currently quite small and they represent only a medium-term competitive
threat.

                             MAJOR RECENT TRENDS IN
                        MALAWI'S TOBACCO SUBSECTOR

Production Trends


        Table 3 depicts the trend in tobacco production in Malawi since the mid-1990s.
Following a severe drought in 1994, production recovered and continued to expand until
reaching a peak in 1997. Since then production has leveled off, with year-to �year variations
affected by the weather, prices, and the financial accessibility of farmers to fertilizer (either
through savings, credit, or the diversion of free/subsidized fertilizer from food crops to tobacco
production). Indeed, with the relaxation or unenforceability of restrictions on tobacco production,
the above factors �together with the more general declining viability of much of the estate sub-
sector--have largely driven tobacco output.

        The official production figure for 2001 represents an undercount, probably of a minimum
of 5000 tons and perhaps between ten and fifteen thousand tons. There is evidence that some
significant quantity of tobacco--at least 5000 tons--was sold by farmers into neighboring
countries and then re-imported into Malawi as `Zambian' or `Mozambican' tobacco in order to
avoid the payment of Malawi auctioning fees and other institutional levies and in order to avoid



                                                 12

the overall logistical problems and transaction costs associated with conducting sales through the
officially-recognized auction-linked channels.21 Taking these developments into account, overall
tobacco production in Malawi in 2001 probably matched the levels of 1998 and 1999, yet fell
short of the two peak years of production which were 1997 and 2000.

        The rapid growth in Malawi's tobacco production which characterized the 1980s and first
half of the 1990s has since given way to a leveling off, with year-to-year fluctuations. Production
trends, however, vary significantly among different varieties of tobacco. Since the early 1990s,
there has been a large and steady decline in the country's production of flue-cured tobacco, with
this decline being most dramatic over the past two years. This crop is grown primarily by
medium and especially large-scale estates and estate companies. Sharply reduced prices in the
late 1990s rendered many such operations unviable, plus there have been additional concerns
(and costs) associated with the diminishing availability of wood to fuel the curing process for
this variety of tobacco. Several long-standing flue-cured tobacco growers ceased production
altogether while others reduced their plantings.22 Currently, a belated effort is being made to
revive Malawi's flue-cured tobacco production, based on finance which was internationally
sourced and guaranteed by Malawi's leading tobacco buyers.

                         Table 3: Malawi Tobacco Production Trends (OOO Tons)

    Year                  Burley                 Flue-Cured              Western Types     Total
    1990-93 Ave.          85.3                   27.3                    11.9              124.5
    1994                  71.3                   20.7                    5.5               97.5
    1995                  101.4                  19.9                    8.8               130.1
    1996                  117.9                  15.4                    8.3               141.6
    1997                  133.9                  14.9                    9.3               158.1
    1998                  113.8                  13.9                    6.7               134.4
    1999                  111.4                  14.3                    8.8               134.5
    2000                  142.2                  10.7                    6.8               159.7
    2001                  115.3*                 8.3                     1.0               124.6*
    2002                  125.4                  11.2                    1.6               138.2
    * Does not include `diverted' crop, estimated at 5000 to 10,000 tons.
    Source: Tobacco Control Commission

        Also underpinning the wider trends has been a stagnation and more recent virtual
collapse in production and sales of various so-called Western types of tobacco (i.e. dark-fired
and sun-aired) with reduced support provided by the parastatal ADMARC, the historical
promoter of these crops. The Malawi tobacco industry has increasingly become dominated by the
burley crop, the output of which continued to rise until 1997 and has leveled off since then
within a moderate range. Still, taking into account estimates of the `diverted' burley crop in
2001, the volume of burley production was higher last year than in all years other than 1997 and
2000.




21There is some suspicion, although no clear evidence, that some additional tobacco crop was sold directly by
growers to the international buyers who operate in Malawi. Under current regulations this would have been illegal.
22See the discussion on tobacco profitability for estates in Olney et al. (1998).



                                                          13

Continued Structural Change


         The significant change in the structure of burley production which began in the early to
mid-1990s has continued in recent years. Smallholder farmers--defined as those who own less
than two (and normally one) hectares of land and generally cultivate only 0.1 to 0.3 hectares of
tobacco--now account for some 70% of the total national tobacco crop. The share of the national
crop of both small/medium estates and large estates has declined substantially and probably
consistently since the mid-90s. (Table 4).23

         Many estates have simply ceased burley tobacco production altogether, having
experienced problems of labor abandonment and the sale of their tobacco by farm tenants and/or
hired estate managers to `intermediate buyers' or others. The tenancy system, which historically
and through to the early 1990s had been a prominent and controversial feature of burley
production, has been widely eroded in recent years and many estate owners have had to abandon
it altogether, shifting to a direct production system with hired labor.24 A substantial number of
estates have experienced more general financial and management problems in recent years.25

         While no hard statistics are available, we estimate that some 40-50% of the estates which
had been producing tobacco in the early 1990s have since given up production altogether, while
many others have scaled back their operations. Absentee-owned estates and larger estates with
relatively high overheads have been most seriously affected by the liberalization of the sector
and the changing economics of tobacco production. Small estates--the operations which had
accounted for most of the growth in the industry during the period from the mid-1980s to the
early 90s-- have also experienced a decline in recent years with many losing their regular access
to credit, either due to their own loan defaults or to the general reductions in agricultural lending
by Malawi's commercial banks (see below).




23This analysis is based on the author's estimates, although these are consistent with the smallholder crop estimates
made annually by the Ministry of Agriculture and Irrigation. Tobacco `production' data provided by the Tobacco
Control Commission no longer reflect actual patterns as a large proportion of tobacco sales by registered estates is
now of tobacco which they have purchased from smallholders or others, rather than directly produced themselves.
The breakdown in the integrity of the registration system for tobacco production has undermined the reliability or
meaning of official statistics.
24Similar to that which has always prevailed for flue-cured tobacco production in Malawi.
25Recent empirical analysis of Malawi's estate sub-sector is lacking. The last substantial and quantitatively based
piece of research on the estate sub-sector was the Estate Land Utilization Study of 1997. A more qualitative review
of the `estate crisis' was provided in 1998 by Olney et. al.



                                                          14

               Table 4: Structural Transformation of Malawi Burley Tobacco Production
                                               (Author's Estimates)

           Total      Smallholder         Small     Medium      Smallholder       Small     Medium/Large
         National      Production        Estates + Larger        Share (%)        Estate     Estate Share
           Crop        (000 Tons)         (000       Estates                      Share            (%)
           (000                           Tons)       (000                         (%)
           Tons)                                     Tons)
1994         71             11             30          30              16            42             42
1995        101             24             36          41              24            36             40
1996        118             46             37          35              39            31             30
1997        134             72             36          26              54            27             19
1998        114             73             23          18              64            20             16
1999        111             76             20          15              68            18             14
2000        142             94             25          23              66            18             16
2001       125**            88             20          17              70            16             14

    *Estimates based upon smallholder club numbers, MOAI smallholder crop estimates, IB sales data,
    estimates of estate IB activity, etc. Between 1994 and 1998, it is estimated that 85% of sales by
    intermediate buyers was of smallholder production, the balance being crop sold by estate managers and
    tenants. From 1999 onwards 95% of the IB sales are assumed to be of smallholder origin. The
    proportion of the recorded `estate' sales in the auctions which originated on smallholder farms is
    estimated to have risen from 10% in 1994, 15% in 1995, 30% in 1996, 405 in 1997, and 50% thereafter.
    All smallholder club sales are assumed to be smallholder origin crop, except in 2000 when an estimated
    5% was of estates who registered as clubs to escape the withholding tax. All the resulting estimates for
    smallholder production are within 10%, plus or minus, from MOAI smallholder burley crop estimates.

    ** Includes some 10,000 tons which may have been sold across Malawi's borders illegally.


         The number of smallholder farmers who are engaged in tobacco production seems to
have reached a plateau after rising rapidly in the mid to late 1990s. It was previously estimated to
have risen from 50,000 in 1994 to between 175,000 and 200,000 smallholders in 1996.26 . The
1997/98 Integrated Household Survey found that some 18.9% of smallholder households--
equivalent to some 359,100--- were cultivating tobacco. Participation has since leveled off and
probably declined somewhat.27

         In 2001, there were some 23,363 registered burley clubs although only 19,714 actually
sold their tobacco on the auction floors. Tobacco Control Commission registration figures
suggest an average of fifteen members per club. Assuming that some proportion (perhaps 5 to
10%) of registered clubs are fictitious or one/two person clubs (registered by estate owners,
agricultural extension officers, or others), one can estimate that some 315,000 to 330,000
smallholders currently are producing tobacco. According to TCC data, just under one-third of
registered tobacco club members are women.


26Jaffee 1997
27Some proportion of smallholder tobacco growers have shifted, at least in part, to the cultivation of paprika. This is
also a high-value crop which derives from the same plant family as tobacco. According to the major exporting
company some 75,000 farmers are growing paprika in the 2001/02 season, up from 40,000 the previous year



                                                          15

         In the mid-to-late-1990s, continued smallholder entry was facilitated by a number of
factors. One of these was strong economic incentives. At least during that period, tobacco
cultivation was generally far more renumerative than most other crops grown by smallholders,
plus the marketing system for this crop was far better developed than it was for virtually all other
agricultural commodities.28           A second important factor was the broad spread of technical
knowledge about (at least aspects of) tobacco cultivation. Hundreds of thousands of rural
Malawians had prior experience with tobacco--serving as laborers or tenants on estates,
cultivating dark-fired or other `Western' tobacco varieties, or cultivating burley tobacco illegally
prior to the reforms.

         Further facilitating the spread of smallholder burley production were three areas of
institutional development. One involved the formation of `burley clubs' both with the support of
the Ministry of Agriculture and under the auspices of the USAID-supported Smallholder
Agribusiness Development Project. These clubs--whose numbers grew from a few thousand to
more than twenty thousand between the mid-1990s and 2000, have variously served to assist
smallholder tobacco growers in procuring inputs, accessing credit, accessing technical assistance,
and organizing the transport of their tobacco to the auction floors. The SADP has given rise to
the formation of a National Association of Smallholder Farmers of Malawi, a very promising yet
still fragile structure for organized lobbying and service delivery for smallholder farmers.

         The second institutional development was the introduction of a program to license
`intermediate buyers' of tobacco. This complemented the formation/activities of `burley clubs' as
many clubs were (and are) rather weak and as farmers still require an intermediary between
themselves and the (distant) auction floors. These intermediaries provided immediate cash for
harvested/cured tobacco, albeit at highly discounted prices.29 They also provided the logistical
link for farmers who otherwise were not able to bring their tobacco to the auction floors. The
IBs, which numbered at their peak some 4000, essentially competed with estates who both
historically and currently purchased tobacco from smallholders and then sold that tobacco under
their own grower names. The IB program proved to be controversial--and probably accelerated
the demise of the tenancy system-- and was scaled back substantially during the last two seasons.
More on this later.

         The third institutional development was the availability of credit to smallholders,
primarily provided by the Malawi Rural Finance Company.30 This company was established in
1993/94 as an autonomous, though government-owned company. It received technical assistance
and a line of credit under a World Bank project. MRFC was designed to provide loans and other
services to smallholder farmers and small rural businesses. Although not originally anticipated,
the company came to serve as the leading source of finance for smallholder tobacco growers who
had been organized into burley clubs. Some two-thirds of MRFC's loan portfolio in the late
1990s was for inputs for mixed crop production--essentially tobacco and maize--and the
company financed some 30-40% of smallholder burley growers during this period. Those


28Ibid. and Keyser (1996)
29Typically half of the auction floor prices.
30For the most part, Malawi's commercial banks have not financed smallholder agriculture, although periodically
they have attempted small lending schemes with selected crops and growers.



                                                      16

growers who accessed credit typically outperformed other smallholders due to the improved
quality and yield of their crop.

         Significantly, parallel developments did not take place to strengthen the quality and
availability of advisory services provided to the massive number of new entrants into the sector.
The Ministry of Agriculture has a limited number of tobacco specialists, yet the overall field
extension service has experienced problems of mobility and has given most priority to food crop
matters and devoted little attention to tobacco cultivation and post-harvest matters. The center for
research and advisory services for tobacco has been the Agricultural Research and Training Trust
(ARET). While regarded as doing quality work, ARET's long-term focus has been on estate
agriculture and the organization lacks the field presence to substantially impact on widespread
smallholder production. Years of periodic discussions on how to synergize the efforts of the
Ministry and of ARET to bring more effective advisory services to smallholder tobacco growers
have thus far failed to result in a workable strategy and financial arrangements.


Trends in Prices, Quality, and Productivity


         While Malawi's overall tobacco production has leveled off in recent years, the prices
realized on its auction floors have declined sharply. This can be seen from the graph below. Both
for burley and flue-cured tobacco, auction prices peaked in 1996, giving rise to the large (burley)
production response the following year. For burley, prices subsequently fell, recovered
somewhat, and then fell very substantially such that average prices in the past two seasons have
been nearly one-third lower than those which prevailed in 1996-1997. For flue-cured tobacco,
prices fell substantially from 1996 to 1998 and then leveled off at this lower position. For the
2001 season incentive prices were paid to flue-cured growers as the international buyers feared
the total collapse of Malawi production of this crop--at the same time when their supplies from
Zimbabwe have become extremely vulnerable to decline.

         This movement in the auction prices in Malawi parallels the wider trend in the
international market. At least in those countries where producer prices for tobacco are not
affected by substantial price supports by government, there occurred a substantial erosion of
producer prices between the mid-1990s and 2001. This can be seen in the graph below. The
erosion in prices has been proportionally even greater in Brazil than it has been in Malawi.
Structural imbalances in world supply and demand for burley tobacco--and the related build-up
of unsold stocks and the need to find new (frequently lower priced) market outlets for available
product--are the proximate cause for these general price developments.31




31Another contributing factor may have been the uncertainty, faced by major cigarette manufacturers, regarding the
extent of legal litigation which they would be facing as well as the size of the financial settlements which would be
agreed to. Even though tobacco leaf itself accounts for a very small proportion of the overall costs and prices of
cigarettes, these developments likely led the manufacturers and the tobacco brokers which supply them, to seek to
reduce the costs of raw materials.



                                                          17

                                  Figure 2: Recent Trends in Average Tobacco Auction Prices in Malawi

                               2.25

                                 2
                      Kg.
                               1.75
                         per
                            $  1.5
                             US
                               1.25

                                 1
                                     1994    1995    1996     1997    1998    1999    2000     2001    2002

                                                                Burley      Flue-Cured



                                              Figure 3: Producer Prices for Burley Tobacco


                               2.5

                                 2

             Kg.                                                                                   Malawi
                               1.5                                                                 Brazil
                Per
                                 1                                                                 Zimbabwe
                   US$                                                                             India
                               0.5

                                 0
                                    1995     1996     1997    1998     1999     2000    2001




       Somewhat exacerbating these general (global) price trends have been changes in the
composition of the Malawi burley tobacco crop since the mid-1990s. While empirically
demonstrating these changes is not possible with available data, there is a widely expressed view
among industry stakeholders in Malawi that the overall quality of the burley crop has been
somewhat reduced and that certain styles and grades of tobacco are no longer available in the
desired quantities. While the international buyers indicate a need for a `balanced' crop,
comprising a wide spectrum of tobacco styles and grades--each with their own uses and market
outlets--there is an expressed view that in some years the composition of the crop has not fully
complied with market requirements and opportunities.

       Whether this is in fact true or merely a convenient bargaining tool--by the buyers to
rationalize lower prices and even by Malawi's estate interests to heap scorn on the validity of the
subsector reforms and the emerging dominance of smallholder farmers--is difficult to discern.




                                                                  18

There are a number of factors which bring plausibility to the notion of quality reduction in
Malawi's tobacco crop since the mid-to-late 1990s. These include:

    � the rising proportion of the crop which is accounted for by highly undercapitalized
         smallholder farmers, typically growing the crop with own saved seed and with little or no
         fertilizer,
    � growing environmental constraints, including the reduced availability of poles and thatch
         for proper curing barn construction;
    � the entry and subsequent exit of large numbers of intermediate buyers, some proportion
         of which lacked prior tobacco experience and/or suitable facilities to grade and bale the
         purchased crop,
    � continued logistical problems for the delivery of smallholder tobacco from rural areas to
         the auctions (based in Lilongwe, Mzuzu, and Limbe) and major problems of `congestion'
         associated with deliveries to the Lilongwe auction; and

    � a lack of strong financial incentives for farmers to produce a high quality crop, given the
         small premiums paid on Malawi's auctions during the mid-to-late 90s for the higher
         grade tobacco compared with the average or below average quality of tobacco.32

         The only quality classification system that really counts is that of the international buyers
of Malawi's tobacco. These classifications, however, seem to be regarded as something of a
trade secret, and the underlying data on the quality mix of the Malawi crop is not available in the
public domain. The only quality classification system for which data are publicly available is that
conducted each year by the Tobacco Association of Malawi (TAMA), an association
representing (primarily estate) growers. TAMA's tobacco classifiers operate on the auction
floors, moving along the lines of bales, taking quick samples and assigning a style and quality
grade to each bale. This system, although highly subjective and subject to a margin of error, is
viewed as providing some value to growers in helping them relate auction prices to different
types and qualities of tobacco leaf.

         Time series data from the TAMA classification system (See Annex A) do not depict a
generalized pattern of declining quality among Malawi's burley crop. Instead, those data point
to: (i) a slight decline in the proportion of the crop classified as 1st or 2nd grade (from 28% in
1995 and 1997 to 21-24% in the past two years), (ii) considerable stability in the proportion of
the crop classified as 4th or 5th grade (generally around 20%), (iii) a small increase in the share of
the crop given the middle 3rd grade (now some 53-55%), (iv) a general improvement through the
long-term reduction in the proportion of the crop which is classified as mouldy, and (v) an
atypical situation for the 1999/2000 season crop which was qualitatively inferior, compared with
the general pattern since the mid-1990s.



32For example, in 1997 the average auction price for burley tobacco was $1.53/kg.. The premium for 1st and 2nd
grade tobacco was negligible (i.e. only $0.05) while the reduction for low grade or even moldy tobacco was not
substantial. Moldy tobacco, together with 4th and 5th grade tobacco earned the farmer $1.29/kg. in that season. In the
past two seasons, the international buyers have moved to more strongly reward high quality and penalize lower
quality tobacco through their price offers in the auctions.



                                                          19

        Over an extended period, Malawi's tobacco farmers has been experiencing a decline in
productivity. According to the data of the Tobacco Control Commission, Malawi's average
yields for burley tobacco production have fallen, more or less steadily, from 1150 kilograms per
hectare in 1990 to 922 kilograms per hectare in 2001. Average yields have also fallen for flue-
cured production from 1760 kg/ha in 1990 to 973 kg/ha in 2001. For these varieties of tobacco,
such yields are extremely low by international standards, and, in recent years, approximately one
half or less than the yields obtained in each of the major tobacco-producing countries with whom
Malawi competes. This comparison of yields is summarized in the following graph.


                             Figure 4: Comparative Tobacco Yields (KG/HA)




     2,500
                                                                                   United States

                                                                                   Zimbabwe
     2,000
                                                                                   China

                                                                                   Brazil
     1,500
                                                                                   Argentina

                                                                                   India
     1,000                                                                         Malawi*


       500
             1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000


    Among the factors which have likely contributed to this decline in Malawi tobacco yields
include:
    � The decline in production or absolute exit from the sub-sector by many of the long-
        standing estates,
    � Lack of adequate advisory services to smallholder farmers,

    � Low and probably declining use of fertilizer and a more general decline in soil fertility,
    � Widespread use of low-quality, own-saved seed,

    � An increased incidence of disease and pest infestation due to inadequate crop rotations,
        and
    � Significant post-harvest losses because of inadequate curing barn infrastructure,
        especially although not exclusively on smallholder farms.


Declining Profitability


        In the face of sharply reduced prices and what have most likely been adverse movements
in product quality and yields, the profitability of tobacco cultivation in Malawi has been sharply



                                                  20

reduced. While the contraction of profits first had a major adverse effect on the viability of estate
production, the trends in the past two seasons have resulted in very marginal returns for
smallholder farmers. An illustration of this is provided in Table 5.

        This analysis is based on crop budgets from members of the National Association of
Smallholder Farmers of Malawi (NASFAM). These farmers, who comprise some 15% of all
smallholder tobacco growers, are more likely than the overall mix of smallholder growers to
access credit, apply fertilizer, and obtain higher yields and somewhat higher prices than the
national norm. These growers also incur somewhat lower logistics and transaction costs than
non-NASFAM member smallholders because of the services and linkages provided by the
association. For the purposes of this analysis we have assumed no change in average yields over
the coverage period, suggesting a somewhat optimistic scenario.

        The analysis suggests that over the 1997 to 1999 period, the net returns per kilo remained
within a range of $0.34 to 0.37 and that these `typical' farmers--with an output of 300 kilograms
of burley--obtained a net income from tobacco of just over $100. While this is not a princely
sum, such a level of cash income was substantial in rural Malawi. According to the 1997/98
Integrated Household Survey the mean cash income33 for all rural households in 1998 was only
$114. That for `non-poor- households was $155, while the mean cash income for `poor'
households was only $71. Tobacco, therefore, continued to play a powerful role in the incomes
of those smallholders who grew the crop.


                  Table 5: Net Returns for (NASFAM) Smallholder Burley Growers

                                     1997          1998          1999       2000     2001
    Ave. Sales Price ($/kg.)         1.56          1.30          1.42       1.05     1.10
    Net Returns/Kg. ($)              0.36          0.37          0.34       0.17     0.09
    Net Returns Per Farmer ($;       108           111           102        51       26
    Output =300 kg.)
    Net Returns Per Hectare          540           555           510        255      130
    Net Returns as % of Ave.         23            28            24         16       8
    Price
    Source: NASFAM data

The table starkly depicts the decline in profitability over the past two seasons, with the returns
per kilogram falling by 50% in 2000 and then almost by half again in 2001. For the latter year,
this implies a net profit for the `typical' NASFAM member of only $26. Among farmers
employing low-input, low output practices it is almost certainly the case that their profits from
tobacco cultivation have declined substantially as well.

        In addition to price, quality, and productivity factors, several other factors have
contributed to the erosion of profitability among Malawi's tobacco growers. One such additional
factor has been the movement (or management) of the exchange rate. In three of the past four
years the pattern of exchange rate movements have been inconsistent with the interests of
tobacco farmers. This is summarized in Table 6.


33Not including the value of home production (of food) which was consumed.



                                                       21

        Purchased inputs for tobacco are obtained just before planting and reflect the exchange
rate at that time. This is especially the case for fertilizer, the most costly purchased input. The
exchange rate in October � November is used here as a proxy for the exchange rate determining
input costs. Most sales of tobacco take place over the June to August months and the average
exchange rate for this period is the proxy for converting dollar earnings at the auctions into
Malawi Kwacha and their distribution to farmers. The three-month period after the auctioning of
the tobacco is a time when much of the tobacco is exported.

        As seen in the table, in both 1997/98 and 1999/00, a devaluation did take place between
the time of planting and the farmer sale of tobacco. This benefited growers. However, in the
three-month period immediately following the farmers' sale there was an even more substantial
devaluation of the Kwacha which provided no benefit for farmers. In fact, that steep devaluation
of the Kwacha just after the closure of the tobacco auctions simply fed into higher input costs for
the farmers during the subsequent seasons. For the most recent season, the Kwacha actually
appreciated during the course of the tobacco growing and selling season, exacerbating the effects
of low auction sales prices.


                  Table 6: Exchange Rate Movements Relevant to Tobacco Earnings

    Season                   Oct-Nov (inputs)      June-Aug (auction)      Sept-Nov (exports)
    1997/98                  17.7                  27.93                   42.0
    1998/99                  42.5                  43.4                    44.5
    1999/00                  45.0                  57.2                    75.1
    2000/01                  79                    71.5                    63.1

        The profitability of tobacco production has been further weakened by the relatively high
and rising costs associated with the transport, handling, and auctioning of Malawi's tobacco.
Tobacco, when transported internally, incurs charges which are well above the road transport
freight rates applied on general goods. While NASFAM has been able to negotiate favorable
rates for its members--typically on the order of US 5 cents/kg. �for transport from the rural
areas to the auctions, most other smallholders--whose tobacco is first sent to satellite depots
managed by TAMA--incur transport costs of US 10 cents or more.

        International transport costs for the exports of Malawi tobacco are also relatively high
since some two-thirds of the exported product is still transported by the long route of road and
rail to Durban in South Africa. Efforts are being made to direct more freight traffic via rail to the
Nacala port in Mozambique, yet there are still technical and management constraints on using
this channel. The cost savings from more extensive use of this export channel might result in
higher prices on the Malawi auctions by some US 7 to 10 cents (that is 7 � 10% of the recent
average prices).

        Malawi growers incur relatively high costs in selling their tobacco over the auction
floors. For example, Auction Holdings Limited, the sole auctioning company, charges growers a
fee of 3.95% of the gross revenue realized. This is very high by international standards. The
comparable rate in Zimbabwe has been 2.4%. The fees charged by AHL to apply a stop-order for
a financial institution are extremely high, being 3% of the loan amount, as compared with the fee


                                                22

of only 0.14% of the loan amount charged under similar circumstances in Zimbabwe. While the
AHL stop-order system has been reasonably effective as a basis for agricultural loan recoveries,
it comes at high cost to growers.34

         In addition to the auction fees, Malawi growers pay a series of institutional cesses which
are also deducted by AHL from the gross revenues obtained by farmers. These include a levies
for ARET (1%), associations (0.85%), TAMA's classification system (0.5%), and the Tobacco
Control Commission (0.13%), as well as a payment of US $0.92 per bale the cover the cost of a
hessian bag replacement scheme. These levies amount to some 3.58% of the farmer's gross
revenues. This level of charges is very high by international standards.35 In past years, the
significance of these levies was relatively modest compared with the earnings potential from
Malawi tobacco. With the recently depressed price structure, the size of these levies, (and the
high fees charged by AHL) need to be re-evaluated with a view toward their downward revision.

         Of additional yet not readily quantifiable significance, are the transaction costs which
many tobacco growers incur which are associated with petty corruption, mismanagement, and
other inefficiencies within the supply chain. Except for the 15-20% of smallholders whose clubs
are members of NASFAM, `direct' sales through the auctions are presently undertaken through a
network of some 88 satellite depots managed by members or counselors of TAMA. In past years,
there were widespread complaints about the mismanagement of such depots (essentially small
warehouses) with `first in-last- out' being practiced with tobacco bales, with bales going missing
or being damaged, and/or with growers having to pay relatively high transport costs for the
deliveries made to the auctions. Growers bore the brunt of these costs as no one was accountable
for lost or damaged tobacco. While some improvements in these arrangements have been
reported, inefficiencies remain and improving returns to farmers is certainly not the driving force
for services provided.

         These transaction costs are then magnified when tobacco from the satellite depots is
delivered to the auction floors. While there has long existed a system of delivery quotas to bring
an even flow of tobacco to the auctions, this system has been overwhelmed in recent years by the
sheer number of growers and trucks conveying tobacco, together with the practice of prominent
and well-connected growers ignoring their assigned delivery schedules and jumping the queue
leading into the auctions. This problem is especially prominent at the Lilongwe auction and lines
of 500 trucks or more are not uncommon during the period from June to August. These trucks
essentially serve as warehouses, holding growers' tobacco in a semi-protected state for up to two
weeks, during which time damage or `leakages' may occur. Farmers wishing to observe the sale
of their tobacco may have to come back and forth to the auction several times as the queue of
delivery trucks moves forward.36

34 Given certain features of agricultural loans in Malawi--especially the need to deposit some 10-20% of the loan
value into the financial institution and thus borrow 110-120% of the amount of one's intended loan--the actual cost
of the stop-over fee to the grower tends to be 4.5% or more, ra ising the cost of finance by this amount.
35 In Zimbabwe, the combined levy for research and for tobacco associations is 0.5%, while the hessian levy is
0.1%.
36 The bottleneck at the Lilongwe auction may have more to do with the processing capacity and use of that capacity
by the international buyers. More tobacco can be handled and sold on a daily basis at that auction than can be
processed in the adjoining factories. In addition, the buyers also use those factories to process tobacco that is
imported from Mozambique and then re-exported.


                                                          23

        While it is not easy to quantify these transaction costs associated with the intermediation
of smallholder tobacco to the ultimate buyers, they are not insubstantial. These inefficiencies and
rents may have been tolerable in a context when tobacco prices were high and grower profit
margins substantial. However, such costs are likely to serve as the difference between rather
small profits and absolute losses by farmers in the context of compressed prices.

        If these marketing and transaction costs were not enough, smallholder growers have
recently been required to pay a 7% withholding tax on the revenues obtained through their
auction sales. This withholding tax is essentially an installment payment for income tax.
Currently, persons earning an income exceeding MK 30,000 are liable for income tax. This
applies to few, if any, of the smallholder tobacco farmers, yet some of their clubs, as an entity,
certainly obtain gross revenues exceeding this sum. For years, smallholders (clubs) were exempt
from the withholding tax, yet this arrangement was soon abused by estate owners who registered
themselves as `clubs', giving a fictitious list of members. With the grower registration system in
disarray, the `solution' was to apply the withholding tax on smallholder clubs, yet make
provision for them to claim back the tax. While the claim form is not exceptionally difficult, it is
apparently very difficult for many of the less well organized clubs (and farmers) to organize their
records and obtain the necessary representation (by an accountant, for example), to actual get
back their withholding tax.

        The combination of statutory levies, high auction fees, high transaction costs within the
supply chain, and the re-imposition of the withholding tax has led a significant number of
growers to by-pass the established marketing channels and seek to sell their tobacco directly to
the international buyers. In several locations on or across the borders to Zambia and
Mozambique, Malawi growers sold their tobacco to the international buyers, their representatives
or others. While paid (in Malawi Kwacha) less than the prevailing auction prices, these farmers
achieved enormous savings in logistics and transaction costs. As noted earlier, probably
somewhere between five and ten percent of the burley tobacco crop was sold through these
extra-legal channels in 2001. This represents something of a warning: a flashing yellow light that
the present system of logistics and marketing for the smallholder crop entails intolerable risks
and costs for many of the smallholder participants.

        An additional `strategy' employed by farmers to cope with the cost-price squeeze is to
default on their agricultural loans, in part or in full. Credit repayment rates for tobacco growers
have declined since the mid-1990s. For example, while in the 1996/97 and 1997/98 seasons
MRFC obtained agricultural loan recoveries of around 90%, such loan recoveries have fallen to
about 80% during the past three seasons. Some proportion of loan default involves farmers
which fail to produce enough tobacco (revenue) to pay off the loan. A larger proportion of the
loan default involves farmers whose profitability has declined and whose coping strategy has
been to manipulate the tobacco registration system, obtaining multiple grower or seller numbers,
and then selling the tobacco under the name(s) for which there are no stop-orders placed for their
loans.

        This process, occurring on a significant scale, has threatened the sustainability of
agricultural/rural finance in Malawi and led to an array of ad hoc schemes by government to
intervene in the credit market, including through the provision of credit repayment guarantees to



                                                  24

input companies. The credit repayment experiences of these government-backed schemes has
been very poor, further eroding credit discipline in the sector.37 More generally, the breakdown
in credit discipline has resulted in a contraction in agricultural lending by Malawi's financial
institutions. This is illustrated in the graph below which combines the agricultural lending of the
commercial banks and of MRFC and deflates these by the CPI. For Malawi's commercial banks,
agricultural lending declined by some 75% in real terms between 1990 and 2000. The increased
lending of the Malawi Rural Finance Company has only partially cushioned the blow from this
contraction in commercial bank lending to the sector.

          To a considerable extent, a vicious circle has been formed in which low profitability
leads to willful (or non-willful) loan default, resulting in lost access to commercially provided
finance. This has led some farmers to reduce their fertilizer use and cut other corners in
production, thereby reducing their yields and/or quality and the profitability of their crop. Those
farmers lucky enough to gain access to the latest government credit scheme, generally under the
false pretext that either they have not previously borrowed or are not a `defaulter' can delay the
on-set of this cycle by a year or two until they default again or the latest scheme collapses under
the weight of massive default.

                                      Figure 5: Commercial Lending for Agricultural in Real Terms

                                     180
                                     160
                                     140
                                     120
                                     100
                         Kwacha)     80
                                     60
                                (!990
                                     40
                                     20
                                      0
                                         1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000




Production Cost Economies


          While the decline in grower profitability over the past three to five years is very
significant, that the crop remains profitable at all can be attributed to a reduction in production
costs, amounting to some 25% between 1997 and 2001. Part of this cost reduction has been for
fertilizer costs, reflecting, especially, the decline in international prices for urea.38 The costs of



37The rate of loan default on the 1998 and 1999 `Agro-Input Schemes' was 25% and 34%, respectively.
38Still, fertilizer costs in Malawi are considerably higher than those in neighboring countries as a result of higher
transaction costs, hedging against exchange rate uncertainty, limited competition, and other factors. In 1999, the
wholesale prices for urea and CAN were 40-50% lower in Zambia and Tanzania than in Malawi. The wholesale
price for urea in Zimbabwe was only one-third that of Malawi in that same year.



                                                                 25

various farm implements have also declined in US$ terms due to the increased volume of local
sales and new entry and competition in the production and sale of such implements.

        Labor costs in Kwacha terms have not kept pace with the devaluation of the currency,
indicating that farm labor has borne a considerable amount of the cost of adjustment to declining
tobacco prices. A serious reduction in labor costs seems to have taken place between 1998 and
1999, while labor costs have remained virtually steady (in US$ terms) since then. This
development needs to be examined further as it has significant implications for poverty in rural
Malawi. The graph below illustrates these trends, indicating labor and purchased input costs per
hectare since 1997.

                      Figure 6:Smallholder Burley Production Costs (US$ per Hectare)



                   1400
                   1200
                   1000                                                       PurchasedInputs

                    800                                                       LaborCosts

                    600                                                       TotalProductionCost
                    400
                    200
                           1997      1998      1999      2000       2001



        Direct comparisons in production costs between Malawi and other countries is limited by
data availability and by the fact that different growers--especially in Malawi--are characterized
by wide variations in their input use and productivity. For the comparison below we utilize the
NASFAM crop budget for smallholders (assuming higher input use and much higher yields than
the norm among such growers) and data from a large company with many estates which still
produce burley tobacco.

                      Table 7: Comparative Production Costs for Burley Tobacco

               Country                     Cost/Hectare (US$)                Cost/Kg. (US$)
      India                                        749                              0.60
      Brazil                                      1282                              0.70
      Thailand                                    1756                              0.86
      Malawi
      Smallholder (00/01                          1080                              0.72
      Large Estate (99/00)                        2062                              1.29
      United States                               8850                              4.14
    Sources: USDA Attache Reports; ITGA Annual General Meeting; Correspondence; NASFAM Crop
    Budgets; Press Agriculture company data.

        The table suggests that while Malawi's costs per hectare are competitive with those of
other developing countries, its production costs per kilogram are somewhat higher as a result of
its far lower yields. Comparative marketing and logistics costs are not available, yet given the
nature of sales arrangements in many countries--now based largely on direct contracting--these



                                                   26

are expected to be considerably higher in Malawi.39 The table also includes a reference to the
underlying production costs for (high-quality) burley tobacco in the United States. What is
evident is that the continuation of that production has only been possible as a result of support
prices paid by the US government.


Exports and Gross Revenues


        With the leveling off of production, the reduction in auction and export prices, and the
decline in yields, the industry is witnessing a decline in export earnings, overall gross revenues
and returns per hectare of land devoted to the crop. The trends for exports are summarized in the
graph below.

                                          Figure 7: Malawi Tobacco Leaf Exports
                                          MalawiTobaccoLeafExports

                   400,000                                                130,000
                   350,000                                                120,000
                   300,000                                                110,000          Export Value
          (000s)
                   250,000                                                100,000 Tons     Export Volumes
                US$200,000                                                90,000
                   150,000                                                80,000


                         199019911992199319941995199619971998199920002001


        The graph below illustrates recent trends for gross revenues and gross revenues per
hectare of tobacco planted. The bars for gross revenue are the value of the burley and flue-cured
crops on the auction floors in Malawi. These revenues have declined substantially since the mid-
1990s and--with declining profitability--no longer provide the large cash injection into rural
areas which the industry once provided. These figures, divided by the planted area for these
crops provides the line in the graph. Gross revenues per hectare have thus declined from about
$1600 in 1995 and 1996 to just over $1000 in recent years.

        Hence, the overall picture is a very mixed one. Over the past decade Malawi has retained
or even increased its share of world production and trade of tobacco. It has done this despite the
virtual implosion of its estate sub-sector, and in the course of a rapid and dramatic restructuring
of the pattern of tobacco production. This restructuring of production resulted in a significant
redistribution of income, likely a primary factor--through its multiplier effects--of that
economic growth which did take place in Malawi during the mid to late 1990s.




39Additional work is needed to support this argument. However, Brazil's tobacco industry has been experiencing a
mini-boom in recent years despite producer prices and unit production costs both being similar to those prevailing in
Malawi. The primary difference underpinning divergent paths in profitability is therefore likely to be marketing
costs and other institutional deductions.



                                                               27

                      Figure 8: Declining Revenue Indicators: Burley + Flue Cured



         240                                                                          1800
         220                                                                          1600
         200
         180                                                                          1400
         160                                                                          1200
         140
         120                                                                          1000

         100                                                                          800
                 1995      1996       1997     1998        1999       2000      2001

                      Gross Revenue ($ Million)      Gross Revenue Per Hectare ($)


       On the other side of the ledger is a profound concern about the sustainability of the
industry and of many of the institutions developed to govern and support tobacco production,
marketing, and trade. Malawi is now devoting more of its relatively scarce land to a crop which
is yielding less, not only in terms of physical product per area but also in terms of unit and
absolute revenue, export earnings, and investable savings. Many former participants in the
industry have been driven out while many others now find their returns to be quite marginal and
their continued specialization in tobacco simply being a function of the lack of remunerative
alternatives. Our estimates of smallholder tobacco production and profitability suggest that in
each of the years 1997, 1998, and 1999 approximately $26 million was generated in net profits
from tobacco. In 2000, this aggregation of net income fell to $16 million, before falling again to
only $8 million in 2001.

       Malawi's participation in the international market for flue-cured and a variety of
`Western' tobacco varieties has been substantially reduced. The continued reliance upon the
long-standing auction system for marketing Malawi's tobacco is now being questioned while its
associated systems for grower registration and market delivery schedules are being abused to the
point where they create as much uncertainty as proper governance. Other institutions, designed
to support the liberalization of the tobacco subsector, are likewise under attack or under stress. A
breaking point has not yet been reached, yet it is fast approaching.

       In light of the mixed experience and the current plethora of problems facing Malawi's
tobacco subsector, there are some who have argued that the liberalization of burley tobacco
production in the early to mid-1990s was a mistake and that this has brought about a general
decline in the industry, via an `uncontrolled' expansion in production, a reduction in the quality
of the crop, and resultant `ruinous' prices which have rendered tobacco cultivation unprofitable
for `serious' producers. This line of argument confuses certain causes and effects. And, it
assumes that the maintenance of the old system of production control was both politically and
economically feasible, a doubtful proposition following the move to multi-party elections in the
early 1990s. On the contrary, a rather strong case can be made that Malawi and its tobacco


                                                  28

industry would have been in a far weaker position today had the tobacco industry reforms not
taken place when they did during the early to mid-1990s. (see Text Box 1)

        Rather than an uncontrolled expansion of tobacco, what occurred in the late 1990s was
simply the redistribution of production from estates to smallholders and, on a more micro-level,
from smallholder households growing tobacco as tenants on estates to smallholder households
growing the crop on their own behalf on their own land. Production expanded to the level (or,
more accurately, to the range) of output which the international buyers have indicated is the level
of demand for the Malawi product.


                      MAJOR ISSUES AND POLICY OPTIONS

        The above trends of declining revenues, prices, and productivity have simultaneously
created a sense of gloom about the role and prospects for the industry among policy-makers and
perhaps the general public, while intensifying the dialogue among stakeholders within the
industry and the search for solutions. This section explores some of the key questions and
options facing policy-makers and industry stakeholders.


Diversify From a Strong Base


        Malawi is highly dependent upon tobacco for its foreign exchange earnings. This
dependence is dramatically higher than is the case for the world's other major tobacco producing
and exporting countries. Table 8 brings out this stark contrast. This high dependency, coupled
with perceptions that tobacco is somehow a sunset industry in the face of the WHO's Framework
Convention on Tobacco Control, has set off alarm bells and urgent calls, both from within
Malawi's government and from outside to accelerate the diversification of the country's
economy.


               Table 8: Leading Tobacco Leaf Exporters: Shares of National Exports

             Tobacco       All     Leaf X as  Leaf X as %       Tobacco     Leaf X as % of      Tobacco
               Leaf      Tobacco     % of        of Total     Product X as       Total         Product X as
             Exports     Product   Tobacco    Agricultural     % of Total     Merchandise       % of Total
            ($Million)   Exports   Product      Exports       Agricultural      Exports        Merchandise
                        ($Million)    X                         Exports                          Exports
USA            1301       5198        25            2              10             <1               <1
Brazil         893         961        93            7              7               2                2
Turkey         479         562        85            11             13              2                2
Zimbabwe       426         444        96            51             53             20               21
Greece         362         503        72            12             17              4                5
Malawi         274         274        100           75             75             57               57
India          215         260        83            5              6              <1               <1
China          209         338        62            2              3              <1               <1
Italy          185         194        95            1              1              <1               <1
 Argentina     166         197        84            2              2              <1               <1
Source: FAO Trade Database




                                               29

                             Box 1: Malawi without Burley Liberalization


 What might have happened had burley liberalization not taken place in the early 1990s? There is
 some (limited) sentiment that things would have been better--both for the competitiveness of the
 industry itself and for the wider economy-- had the burley subsector not been liberalized and
 licensed production still restricted to `qualified growers' (i.e. estates). Recall, however, the
 situation in the late 1980s and early 1990s when the subsectoral growth which had taken place
 was almost entirely among small-scale estates--essentially larger smallholdings-- plus a boost in
 illegal production amongst communal area smallholders who sold through estates (Mkandawire et
 al. 1990). The tenancy system was under strain and problems of labor and financial management
 were severe on many estates, especially medium and large scale estates whose owners were
 weekend farmers (Ibid.; ELUS (1997). Given that situation, together with the subsequent large
 reduction in international prices for tobacco, one can envision that in the absence of the burley
 reforms, the following types of developments might have taken place:

     �    continued growth in production of illegal tobacco, grown largely without fertilizer or
          technical support and thus a significant deterioration of the overall quality of the crop;
     �    widespread labor unrest on the estates and the possible spread to a wider destabilization of
          the economy;
     �    the failure to emerge of a viable smallholder club and association structure which did arise
          on the backs of the burley liberalization;
What�would have happened had not the liberalization of the market taken place in the burley
          the failure to emerge or sustain organizations such as MRFC whose primary customers
subsectorthroughearly 1990s?
           in the  the mid to late 1990s had been smallholder burley growers;
     �    the likely peaking of burley production around the mid-1990s and a decline thereafter,
          resembling the pattern which did indeed take place for estate flue-cured tobacco;
     �    by the late 1990s and early 2000s, Malawi burley production would probably be in the
          range of 40,000 to 60,000 tons rather than the average of 120,000 which did occur over
          the 1998-2001 period. The loss in sales revenue (over the auction floors) would have been
          in the order of $300 million over this period;
     �    The continued concentration of burley-related incomes would have denied to rural areas
          (and the wider economy) the growth multipliers which did indeed occur as a result of
          putting more money into the pockets of nearly two million people (i.e. 400,000
          households);
     �    There would have been a modest diversification of the economy since some smallholders
          would have shifted to other cash crops, provided that downstream investment in
          processing and marketing would have taken place. However, with the specter of
          widespread labor unrest and with the adverse macroeconomic effects of a substantially
          lower burley crop, it is not obvious that Malawi would have provided an attractive
          investment environment during this period to support such diversification.




                                                     30

                             Box 2: Exit and Voice in Malawi's Tobacco Subsector


     In 1970, Albert Hirschman laid out an interesting formulation on the various approaches which are
taken to respond to the declining performance and services of firms, organizations, and even nation-
states. Such deterioration of performance may either be in absolute terms or relative to that of other
firms, organizations, etc.. Hirschman argued that the most common responses to deteriorating
performance could be categorized into two categories, namely:

     �    exit --i.e. ceasing to buy the product, ceasing membership in an organization, withdrawing
          one's labor from a company, physically migrating) in order to compel changes to correct
          whatever faults led to the exit, or
     �    voice --i.e. expressing dissatisfaction to the management or to some other authority to which
          the management is subordinate in the hope of inducing changes to remedy the problems.

     Exit and voice are portrayed as `impersonations' of economics and politics, respectively. Strategies
of exit and voice may be alternatives or may complement one another. Local and industry
circumstances vary as do the effectiveness of either exit or voice to bring about the needed change. Exit
is an especially viable strategy when there is widespread competition (i.e. in product markets, among
political parties, among hiring firms, etc.). Hirschman notes that exit is more harsh and sometimes less
reversible than voice and if stakeholders are sufficiently convinced that their voice will be effective,
they may well postpone exit. In some cases, therefore, exit will be a reaction of last resort after voice
has failed.

     This conceptual framework is useful in considering the juncture to which Malawi's tobacco
subsector has arrived and in considering the policy and other options which now present themselves. In
the recent past, dialogue within the subsector has largely involved a cacophony of accusations, pointed
at different stakeholders, each perceiving that their interests were significantly in conflict with those of
others. The mechanism of `voice' has been largely ineffective in a situation characterized by power
imbalances, prolific rent-seeking (or more accurately, rent-maintenance), and a government unable to
officiate among competing interests and adjust the support and governance system to meet emerging
needs. In the face of ineffective or muted voice, various forms of exit and/or extra-legal activity have
been deployed by stakeholders in order to secure whatever benefits one can from a system whose
profits have been contracting.

     Attempts at `voice' have taken various forms --position papers, letters to government officials
and/or donor agencies, interviews in the media, private meetings or letters written to various marketing
and financial institutions, etc. This `voice' has generally been ineffective, resulting both in frustration
among stakeholders and a general inability to yield institutional improvements which contribute to the
competitiveness of the subsector. Levels of frustration have grown to the point where during the past
three seasons there have been protests and near riots by farmers at the auction floors in the face of what
were perceived as inadequate prices being paid by the international buyers. There have been short
periods where growers have essentially gone on strike, withholding their tobacco from sale on the
auction floors.

     Frustrated by the ineffectiveness of `voice', some stakeholders are pursuing or threatening to
pursue the option of `exit'. Large numbers of growers have ceased to produce the crop. This initially
took place among estates, but there is also evidence of smallholders now shifting away from tobacco.
This may in fact be healthy for the industry on a longer term basis. Other forms of `exit' may be less
sanguine for the sustainability of the industry. One is the strategy of stealth production, obtaining
numerous grower registration numbers and then directing sales to the auction in a manner which evades
loan repayment, tax and levy payments, or other responsibilities. Another partial form of `exit' emerged
last season with the extra-legal cross-border sale of tobacco to avoid the Malawi auctions and
associated transaction and other costs




                                                         31

         There is certainly a need to foster the diversification of the Malawi economy and to
encourage new forms of specialization, especially in areas which are labor intensive and can
therefore offer the prospect for broad-based growth. Given the small size of the Malawi market
much of this drive for growth and diversification will have to be export-oriented, focused on
servicing the Southern Africa regional market as well as competing internationally. Trade
agreements within COMESA and SADC as well as the U.S. AGOA initiative and the Everything
But Arms initiative of the EU may provide some additional market opportunities to Malawi's
would-be investors and exporters. However, this will be a medium to long term process,
requiring improvements in infrastructure and the business environment, investments in human
capital, and the attraction of foreign direct investment or other types of strategic partnerships.40

         Keeping in mind the longer term nature of the process of economic diversification and
integration, Malawi must--in the near term, move to strengthen the competitiveness, improve
the stream of incomes, and enhance the sustainability of its dominant tobacco subsector. Incomes
from tobacco will continue to be important in generating savings and in providing some
additional purchasing power within the domestic market. Those resources already being put to
use in tobacco cultivation and marketing must be made to yield a higher return by achieving
productivity gains and by radically reducing transaction costs within the tobacco supply chain.

         In the near term, Malawi's most prominent engine of growth simply cannot be allowed to
stall. The international market environment may well prove to be more challenging than in the
pinnacle years of Malawi's industry. Future price prospects are probably not good. Still,
Malawi's burley tobacco is well placed in the international market and there is certainly the
potential to achieve a modest revival of the country's flue-cured tobacco production and exports.
Even if the Framework Convention turns out to be very strong, the world tobacco market is not
simply going to wither away and there will be significant demand for tobacco products for many
years (decades).

         Looking forward, it is unlikely that the country will ever achieve the value of tobacco
exports which were recorded in 1997. Nevertheless, if the industry can achieve a partial recovery
of flue-cured production (say to 15,000 tons) and generally meet the international demand for its
styles and quality of burley tobacco (say in the range of 120,000 to 140,000 tons of green leaf
equivalent), the country will likely be in a position to maintain a level of tobacco exports in the
range of $250 to 300 million per year.




40An upcoming study on "The Integration of Malawi into the World Trading System" will be exploring an array of
policy, trade facilitation, and market access issues for Malawi and lay out a program of needed capacity building
within the country to improve the country's trade performance and its impact on poverty.



                                                          32

                      Box 3: SWOT Analysis for Malawi's Tobacco Subsector

            Strengths                                   Weaknesses

            Political stability                         Macroeconomic instability
            Established customer base                   Land/population pressures
            Experienced farmers and workers             Soil infertility
            Modern processing facilities                Weak credit repayment discipline
            Good market information                     High transportation costs
            Price and cost competitive                  Absence of domestic market
            Biggest exporter of burley tobacco          Vested interests against change

            Opportunities                               Threats

            Enhance productivity and quality            Pest and disease build up
            Preference for low nicotine tobacco Lack of wood for curing
            Revive flue-cured tobacco output            Future erosion of MRFC
            Instability in Zimbabwe                     Big expansion in Mozambique
            Reduce transport cost via Nacala            Brazil expands and permanently
            Reduce marketing/transaction costs fills flue-cured market gap
            Improve subsector dialogue                  International health concerns



Core Agenda Summarized

        Whether it can do this and whether such exports can again spur growth multipliers and
contribute once again to rural poverty reduction will depend upon effective improvements in
support systems and in the efficiency of the supply chain, together with more general
improvements in infrastructure and in economic management which lay beyond the scope of the
tobacco sub-sector itself. The core agenda to improve the viability and sustainability of Malawi's
tobacco sub-sector would include the following elements:


Transparency and Economic Stability

    � Considerably increase the transparency in the structure and performance of the tobacco
        sub-sector, starting with the revamping of the grower registration system and putting in
        place an annual auditing and performance review system for the tobacco supply chain
        and each of its key constituent parts (including farmer clubs, satellite depots, intermediate
        buyer systems, the auction system, and the set of exporting operations).41 Part of this
        effort will require strengthening the statistical and monitoring capacities of the Tobacco
        Control Commission42;

    � Improve macroeconomic management, leading to a more predictable movement of the
        exchange rate and contributing to the reduction of interest rates;

41Maintaining the confidentiality of certain private information need not hamper an overall drive to increase
transparency in the industry
42As well as balancing the representation on the board of the TCC to include more grower representation.



                                                          33

    � Introduce a national identity card system which, in addition to its more general benefits,
        will help to underpin the revamped tobacco grower registration system and improve the
        management of agricultural credit. For the latter, an additional measure would be the
        development of a credit reporting system which has been long discussed but not acted
        upon.


Cost Reduction and Productivity Improvement

    � Continued investments and management improvements within the Nacala corridor rail
        and port system, leading to a reduction in the transport costs both for imported inputs
        (e.g. fertilizer) and for Malawi's tobacco and other agricultural exports;
    � Measures to drive down production costs, primarily the cost of fertilizer. This will require
        more research to devise fertilizer recommendations based on economic as well as
        technical factors plus other measures to reduce risk and increase the efficiency of the
        fertilizer distribution system43;
    � Measures to drive down financing costs. In addition to the national ID and credit
        reporting systems, this might include the introduction of special credit-related judicial
        processes, further capacity-building in revolving credit and savings societies, off-shore
        sourcing of finance by tobacco buyers to on-lend to growers, and `moral suasion' by the
        GOM to reduce the high stop-order fees charged by AHL;
    � Measures to drive down marketing and transaction costs in the tobacco supply chain,
        including government intervention to reduce the auctioning fees charged by AHL, a
        policy change to gradually permit the direct contracting of tobacco between growers and
        buyers, and a drive to professionalize or perhaps outsource the management of a system
        of satellite depots (and/or small decentralized marketplaces)
    � Capacity-building in relation to producer associations, including additional support for
        the extension of the NASFAM system and the possible restructuring of TAMA and its
        merger with other associations to bring about a commercial farmer producer organization

        The above set of initiatives will require a combination of government and private
stakeholder action. They may also require the technical and financial assistance of the
international development community and an active role by the latter to break the inertia which
seems to have paralyzed the industry from taking the necessary steps to adjust to change and re-
chart the industries future with a coherent strategic vision.

        Below we elaborate on some of the recommended measures outlined above. Before doing
so it is important to critically evaluate one policy option which has been embraced by some
subsector stakeholders and which government appears to be moving to adopt. This is the policy
option of changing the regulatory framework to allow (presumably licensed) tobacco growers to
export their tobacco (green) leaf themselves and thereby bypass the local auctions and the
international buyers who purchase the crop on those auctions. The premise behind this policy
option is that Malawi growers are not being paid the proper or competitive price for their crop


43See Jungbluth and Hiwa (1999) and Westlake (1999) for a discussion of these measures.



                                                    34

and that profitability could be restored, on a more sustained basis, by by-passing existing buyers
and finding alternative buyers or intermediaries with the international market.


Allowing Direct Exports by Growers: Is the Grass Greener on the Other Side?


        There are long-standing concerns in Malawi that growers are not paid a fair or
competitive price because the purchase of Malawian tobacco is dominated by three international
companies--affiliates of Dimon, Standard Commercial, and Universal Leaf, all of the United
States. These three companies account for well over 90% of the leaf purchases made at Malawi's
auctions, the balance being purchased by a number of smaller private companies and the
parastatal ADMARC. It is alleged that the `share-out' arrangement among the three main buyers
constrains any competition among them in purchasing the crop. These buyers purchase the
tobacco substantially on behalf of the four or five large multinational corporations that dominate
the international cigarette manufacturing industry. Some two-thirds of sales of Malawi tobacco
are directed to these manufacturers. The balance is sold to an array of smaller manufacturers,
including those based in `secondary' markets such as Egypt, Eastern Europe, etc.

        Because of the wide variations in the tobacco crop, it is difficult to make direct
comparisons of the prices obtained by growers in different countries. Generally, Zimbabwean
growers have obtained better prices than have Malawian growers for flue-cured tobacco,
although there is a recognition that the Zimbabwean crop is of somewhat higher quality. The
lower transportation costs of moving Zimbabwean tobacco to southern African ports has also
been a contributing factor.

        For burley tobacco, now the dominant part of the Malawi crop, we showed earlier that in
recent years the `average' prices paid by the international buyers have been similar in Brazil and
Malawi. A similar price pattern and level is also observed for burley tobacco in Argentina,
although the government there has provided price support payments over and above the
prevailing market price. The data available do not enable price comparison between countries on
a grade by grade basis.

        However, given the fact that the Brazilian crop is considered to be of generally higher
quality and value than the Malawi crop, the fact that Malawi auction prices are more or less
similar to those paid in Brazil does not point to any large or particularly unusual `exploitation' of
the Malawi growers. The main differences between the Brazilian and Malawi situations is that
the Brazilians obtain much higher yields (and thus have lower production costs per kg. of
tobacco) and experience lower marketing and transaction costs due to their direct (contracted)
links with buyers.

        There are a couple of other ways of looking at this situation. One is to examine trends
over time in the share of auction payments to growers as a proportion of Malawi's tobacco
export earnings. Intervening variables, such as the carry-over of tobacco stocks from one
calendar year to the next and the changing composition of the country's tobacco crop (among
varieties and grades) put some limitations on year-to-year comparisons. Still, as the graph below
indicates, this proportion has remained within a range of 65-75% since the mid-1990s with the
exception of an outlier year of 1998 when there was an unexpected very large devaluation of the


                                                 35

Kwacha after the tobacco auctions had closed. The share of export values paid out to growers
was essentially the same in 2000 (about 66%) as it was in 1994-95 (68%).44

                   Figure 9: Auction Level Payments as % of Tobacco Export Earnings


           80

           75

           70

           65

           60

           55

           50
                  1994        1995          1996        1997        1998         1999         2000



        An illustrative build-up of value-added of Malawi tobacco from its green leaf stage (sold
at the auction) to its export value is provided in Table 9. A small majority of the purchases of the
international buyers at the Malawi auctions is pre-committed and pre-sold and these prices are
factored into their budgets for their buying operations on the auctions. A profit margin at or
about 15% is sought by the firms.

        Some growers offer the proposition that they could do better by bypassing the
international buyers in Malawi and selling their tobacco themselves to downstream users. They
could either sell green leaf to processors based elsewhere (i.e. South Africa or Zimbabwe) or
obtain custom processing services elsewhere and                        sell  the    processed       tobacco   to
manufacturers/other end users. In response to requests from these larger growers, the Malawi
Cabinet issued a Directive in January 2001 to enable direct grower exports. The Tobacco Control
Commission was instructed to develop the modalities to enable such exports to take place
(including making provisions to collect the statutory levies, for ensuring quality control, and for
ensuring the repatriation of foreign exchange earnings. Also in 2001, one of Malawi's largest
tobacco growing companies applied for and received permission to undertake such direct
exports, although it did not implement such action.




44One cannot make direct comparisons with other countries since the domestic market is a major outlet for most
tobacco-producing countries and so correlations between grower prices and export values are invalid.



                                                      36

                         Table 9: From Auction Floor toOverseas Market (2001):
                            Cost Build-up of Burley Tobacco (US Ce nts/kg.)


             Auction Floor Price                                                    109.77
             Buyer Overhead Costs                                                   11.00
             Short-term Storage Costs                                                2.00
             Threshing Costs                                                        15.00
             Sub-Total                                                             137.77

             Yield (70%)                                                            197.01
             Re-drying and Packaging                                                25.00
             Materials                                                              11.00
             Buyer Profit (10-15%)                                          19.70 � 29.55
             Interest (3 months at 8.5% p.a.)                                        4.95
             Sub-Total                                                    257.66 � 267.51

             Transport to Durban and Port Charges                                   15.00
             Insurance                                                               3.66
             Ocean Freight to Europe                                                18.00
             Cost on Arrival in Europe                                     294.32 � 304.17

             Notes

             Green Leaf Equivalent of Ex-factory Cost                              180.36
             Green Leaf Equivalent of Landed Cost                                  206.02
             *Estimated based on information provided by international tobacco company.

        This `free-market' solution represents a policy option for the GOM and a practical option
for Malawi growers. A number of factors are worth considering. What may be in the interests of
a selected number of growers may not be in the wider national interest or consistent with the
interests of the broader set of small and unorganized growers for whom direct exports--other
than simple cross-border smuggling--is not a viable approach. In examining this policy option,
one should consider a range of possible scenarios.

        Under the hypothetical best case scenario, some of the larger and better organized
growers do in fact arrange for direct sales and/or custom processing and find selected
remunerative outlets for their crop. They save on the local auctioning costs plus also save a
substantial proportion of the intermediary's profit margin. More money flows back into the
pockets of these growers. Malawi experiences a lower value-added on this tobacco (as it is
processed elsewhere) yet its more capable farmers are better enabled to stay in (or even expand)
production and perhaps gain multipliers through their investments. In this best case scenario, this
direct export of green leaf takes some pressure off the local auctions (thereby reducing mid-
season congestion), yet the auctions nonetheless carry on their business. In this scenario, the
international buyers in Malawi seek to win back this directly exported crop by improving the
prices which they pay for the higher grades of tobacco on the auction floors. A dynamic is set in
place in which simply the opportunity to directly export enables Malawi's (better organized)
growers to `bargain' for higher prices in the local auctions. Malawi growers as a whole would be
beneficiaries.



                                                  37

         There are a variety of alternative scenarios which could occur with a regulatory change
allowing direct exporting by growers. One such scenario is that the change gives rise to
widespread opportunistic behavior, with large numbers of growers and traders moving their
tobacco across regional borders, engaging in under-declaring and under-invoicing of such
exports, and more generally contributing to the fragmentation of Malawi's exports and an under-
recording of incomes and foreign exchange earnings. In this scenario, there is the potential for
the further breakdown of existing systems to monitor the quantity and quality of the crop, to
ensure that exports comply with international quality and phytosanitary standards, and to ensure
that growers contribute to the financing of the research and advisory services for the industry. In
this scenario, a greatly reduced volume of tobacco might be sold over the auction floors, perhaps
resulting in under-capacity utilization both there and in the state-of-the-art local processing
facilities. The `integrity' of the Malawian crop--at least in the eyes of the multinational cigarette
manufacturers--could be degraded as supplies of Malawian tobacco are mixed with that of other
countries from within and outside the southern African region.

         Together with or separated from this latter scenario is another possible chain of efforts.
Rather than adopting an accommodating stance to direct grower exports out of Malawi, one
could envision a scenario whereby the international buyers and/or the downstream multinational
cigarette makers adopt a more aggressive posture, threatening or actually undertaking to reduce
their purchases of Malawian tobacco and to replace it with supplies from neighboring countries
or elsewhere.45 A supplemental step would be to sanction the `dissident' crop by pressuring
regional processors not to process the Malawi green leaf or risk losing business with these
international companies. Whether for this reason or because the Malawian growers lack the
range of contacts and logistical skills of the larger international trading companies, it is possible
that the receipt of higher prices by Malawian direct exporter growers is a short-lived
phenomenon.

         Hence, the policy/regulatory framework to permit direct exports by growers entails
certain opportunities as well as certain apparent risks for both growers themselves and the
`industry' as a whole. The downside risks appear to be considerable and in any case one does not
envision a situation where the entire industry, or even one-third of the growers would be in a
position, within the foreseeable future to export tobacco directly. Malawi is a major international
supplier and not a niche market supplier. Eventually, the bulk of its crop will still find its way to
the major multinational cigarette manufacturers. Such manufacturers substantially rely upon
companies such as those who dominate Malawi's current processing and exports for the mix of
tobacco types and varieties which they require. It is unrealistic or even na�ve to believe that the
downstream buyers will undermine the interests of the tobacco merchants simply to save a few
cents on consignments of tobacco out of Malawi.

         Based on this political economy perspective, it is not advisable for Malawi to change its
regulations to enable large numbers of growers to export their (unprocessed) crop directly.


45The prime candidate for this would be Mozambique. That country's tobacco production has increased from very
small quantities to up to 18,000 tons in recent years. A style of burley tobacco similar to that of Malawi's is grown
there on land whose soil fertility is considerably higher. While Mozambique's tobacco has up until now been
processed in Malawi and then re-exported, the feasibility of establishing a processing plant there is being assessed.



                                                           38

Should it be deemed desirable to pilot test such a policy change, then it is recommended that
direct exports be permitted only for flue-cured tobacco, given the much smaller scale of this crop
and the far fewer growers who are active in this crop. Direct exporters would need to be licensed
and a system for monitoring this trade established.


Reducing Marketing, Logistical and Other Transaction Costs


        The system of auctioning tobacco in Malawi at centralized sites dates back to the late
1930s. For many years, this system of sales has served the industry very well and very
effectively accommodated the very large expansion in Malawi's tobacco production during the
1970s and 1980s. It provided a very transparent way for the several hundred, and later, several
thousand tobacco growers to sell their crop.

        Originally, all sales took place at the Limbe auction, just outside of Blantyre in the south
of the country. However, as tobacco production in the center of the country increased, a new
auction facility (and adjoining investments in processing plants by the international buyers) was
put up at Kanengo (just outside of Lilongwe) in 1979. Subsequently, in 1993, an auction floor
was set up in Mzuzu in the north of the country. With the shift in the locus of production to the
central region, the Kanengo auction now accounts for the majority of auction sales.

        According to current regulations (i.e. the Tobacco Act) all tobacco exported from Malawi
must be sold over the auction floors. There is currently only one auctioning company, Auction
Holdings Ltd. (AHL). Over the years, other parties have inquired about the possibility of starting
a new auction, yet none has actually pursued this. While the formal requirements to obtain an
auctioning license are not very difficult to meet, would-be investors have probably been deterred
from doing so by the realization that some significant scale of operation would be needed to have
the international buyers decentralize their purchasing operations. In addition, there has likely
been some official discouragement for new auction investments given the large investments that
the government had already made in the Kanengo and Mzuzu facilities.

        AHL's largest shareholder is ADMARC, a government statutory body. As part of the
country's privatization process, ADMARC's shareholding in AHL has been reduced to 47%.
Part of those shares were divested to TAMA and to individual large tobacco growers, while some
of those shares are held in trust for future purchase by other growers. The governing board of
AHL consists of representatives from ADMARC and TAMA as well as a few larger growers.
Smallholder farmers, who now represent some 70% of the country's tobacco crop are not
represented.

        Thus, there remains a defacto monopoly control of the auctioning function. Smallholder
farmers are not represented in the management of this core institution. Among tobacco growers
there is a widespread perception that AHL primary objective is profit-maximization rather than
promoting the welfare of the growers and supporting the development of the industry as a




                                                  39

whole.46 The relatively high auction fees and the fees charged to handle a credit stop-order
contribute to this perception.

         However, grower dislike and distrust of the `auction system' has as much to do with the
wider supply chain as with the operating functions of AHL itself. AHL can readily claim that this
is not its fault and that many of the inefficiencies and much of the rent-seeking behavior and
petty corruption occurs in the rural areas and far before AHL takes control of the tobacco.
However, the requirements of the `auction system' must certainly change in line with the
structural transformation of tobacco production within the country. While once there were a few
hundred and later a few thousand larger growers who each owned or rented trucks and had their
own staff deliver the tobacco to the auction floors, the production of the crop is now done
primarily by hundreds of thousands of very small farmers, none of whom owns a vehicle nor has
any permanent employees.

         Malawi's auction system is therefore much broader than the auction floors themselves
and the associated sales and payment systems. Part and parcel of this auctioning function is also
the grower registration system and the delivery quota system. The former has been overwhelmed
by the massive new entry into the subsector and its abuse by some people in order to avoid the
payment of levies, taxes, and/or financial debts. The delivery quota system still functions, yet is
ignored by well-connected people and is overwhelmed by the needs of very poor growers to sell
their tobacco as soon as possible in order to obtain cash for basic necessities. The result is
`congestion' at the Kanengo auction and all its attendant costs and impacts on product quality.
The more effective functioning of the auction system requires, at a minimum, the re-vamping of
the grower registration system and the enforcement of the delivery quota arrangement.

         Still, the supply chain for the auction system starts earlier, in the rural areas. At present,
the larger growers continue to deliver their tobacco direct to the auction floors while a minority
of very well organized smallholder clubs, under the umbrella of NASFAM, effectively negotiate
and hire transport services to bring their tobacco to the auctions. The NASFAM clubs have their
own queue leading into the auction floors as do the larger growers. All other growers are
presently required to bring their tobacco to the satellite depots which are located at eighty-eight
sites around the country. These depots are owned and managed by individuals, typically
counselors of TAMA or people selected by them. Farmers or clubs bring tobacco there, it is
temporarily stored and then transported to the auctions. TAMA negotiates the transport rates.

         While a proper evaluation of the satellite depot system has not been done, there have
been widespread complaints among farmers. The performance of those depot operations is
undoubtedly mixed, yet the interests of the growers and those involved in their intermediation
functions do not, necessarily converge. For example, there is little incentive to negotiate the most
favorable transport rates since the profits from such services can be shared between the transport
operations and TAMA. In fact, this past season, NASFAM clubs were paying K350 per bale (75
to 100 kgs.) for transport in the same areas where farmers were being charged K600 per bale by
satellite depots. This extra two to three US cents per kilogram is not insubstantial given current
profit margins.

46In earlier years, AHL retained enough revenue to cover its operating cost and pay a small fixed dividend,
distributing the remainder to growers in the form of a rebate.



                                                         40

        In addition to and probably more problematic than the higher direct charges for
intermediation functions, farmers are concerned about the lack of accountability associated with
the satellite depot system. Periodically, tobacco is damaged or goes missing at the depots or in
transit to the auctions, yet no one is held accountable for this and the farmers affected end the
season without any earnings and perhaps with a defaulted loan.

        Concerns about mismanagement and/or petty corruption in relation to the satellite depots
led some stakeholders to demand that AHL assume the responsibility for managing this stage of
the supply chain and be held accountable for the tobacco once it is delivered to the depot. AHL
was reluctant to take on this function, while TAMA lobbied strongly to maintain its role in the
storage and transport of the crop. While the government seemingly remained agnostic on the
issue, those with the most interest at stake, the poorer and less well organized smallholder
farmers, had no voice in the matter. Inertia prevailed and TAMA has retained this function.

        With the liberalization of the sub-sector, the intermediation between smallholder farmers
and the auctions was supposed to be done by a combination of farmer clubs/groups and
`intermediate buyers'. The development of farmer clubs is still a work in process as Malawi has
no tradition of farmer cooperatives and groupings in the past were at the behest of agricultural
extension officers to reduce the administrative cost of providing advisory services and/or
government-provided agricultural credit. For many years, voluntary group formation and actions
was discouraged (or even forbidden). This picture changed with the political change in the early
1990s. However, many of the former clubs created for extension purposes were not fully
appropriate for joint tobacco-related service activities or for obtaining credit from the MRFC.

        The Smallholder Agribusiness Development Project has made some considerable
headway in bringing a commercial orientation to farmer clubs and in incrementally improving
their management and record keeping. The functions handled by these clubs have been extended
to include collective input procurement, collective transport of tobacco, and, in some cases,
collective marketing of other crops. A stronger understanding about the value and underlying
costs of borrowed finance has been developed among many such club members. This club
structure has, in the past few years, been built upon to form regional groupings and a national
association--NASFAM. Currently, NASFAM has some 93,500 members, including nearly
65,500 members whose lead cash crop is tobacco.

        While the SADP and NASFAM have made impressive gains, these clubs still directly
embrace only a small minority (i.e. about 20%) of Malawi's smallholder tobacco growers and an
even smaller minority (i.e. about 5%) of the country's smallholder growers in general.
Considerably more work is needed, either to expand the membership of NASFAM or to replicate
models similar to it which embrace large numbers of other smallholder growers. The capacities
and performance of the more than 20,000 non-NASFAM clubs which are active in the tobacco
sub-sector have never been subjected to systematic analysis, yet the pattern is certainly a mixed
one, with probably a majority of such clubs displaying major weaknesses which detract from
their services to members.




                                                41

         The intermediate buyer system was introduced in 1994, essentially to supplement the
farmer club system and provide a logistical bridge between smallholders and the auctions.
Intermediate buyers were to buy the cured leaf from smallholders and then transport and put the
crop onto the auction floors under their own registered seller names. It was assumed that only
people with past experience with tobacco would get involved in this function, yet the barriers to
entry were minimal and virtually anyone who paid the licensing fee could enter into the
fraternity of intermediate buyers. No other special provisions were made. Assuming that the IBs
would not be a very significant set of players, they were not required to adhere to a particular
delivery schedule at the auctions.

         The minimal entry requirements and the (lack of) delivery quota loop-hole contributed to
a burgeoning of interest in IB registration. The number of licensed IBs rose rapidly over the 1995
to 1997, peaking in the latter year at just over 4000. Those licensed as IBs included a wide range
of people and entities including small and larger traders, estates, extension agents, civil servants,
and even smallholder tobacco clubs. Motivations for gaining IB licenses varied including
straightforward profit motives, and objectives to bypass delivery quota restrictions, bypass credit
repayment, and other objectives. Some and probably most licensed IBs had prior/on-going
tobacco experience; a majority of IBs didn't have such experience or at least experience in
tobacco grading and marketing.

         As noted in Table 10 below, licensed IBs never accounted for a large share of the crop
sold over the auction floors. The share of licensed IBs in the total national burley crop sales
exceeded 10% only in two seasons--1997 and 1998. The share of IBs--other than ADMARC
which was a major IB operator-- was less than 5% of the crop in every other season during which
the program operated. After 1998, the share of non-ADMARC IBs never reached 3% of the total
market.47

         Intermediate Buyers have been accused of undermining Malawi's tobacco industry,
causing massive theft of the burley crop from grower-owners, undermining the agricultural credit
system, weakening the overall quality of the burley crop through their purchases of unripe crop
and their poor/lack of effective grading. The list of alleged ills is long and the IB program has
seemingly found the spotlight of all concerns and challenges to vested interests which
accompanied the liberalization of the tobacco market, enabling the entry of smallholders into
production.

         Most of the above accusations are inconsistent with available evidence--regardless of the
periodic occurrence of this or that development involving particular IBs. According to the quality
classification system done by TAMA, there is no evidence on a wide scale linking the rise of IB
activity with a reduction in the quality of the crop (See Annex A). Years 1997 and 1998, when
IBs played a prominent role, were totally unexceptional from a quality point of view. The mix of


47In this regard, it is important to note that the licensed IBs initially did and have since represented only a minority
of the entities who in fact have served as intermediaries between smallholder and the auction floors. Prior to the
revision of the Special Crops Act, smallholder farmers grew burley tobacco illegally and sold that tobacco to estates
who subsequently sold the tobacco under their own names and licenses at the auctions. Following the liberalization,
estates have continued to represent the majority of intermediaries between smallholders and the auctions. In many
instances, estate owners had one or more licenses as estates and one or more licenses as IBs.



                                                             42

grades was no different than in the prior two years and the proportion of the crop rated as mouldy
or of mixed grade was actually lower than in the previous years. In 2000 and 2001, once non-
ADMARC IBs had largely ceased operating, the share of the total crop recorded as being of
mixed grade was substantially higher than in all the years that IBS were more active.

         The table below indicates that in most years the average price obtained by licensed IBs
was very close to that of the national average. In most years, the difference was two or three
cents. The biggest differences occurred in 1999 and 2000, yet this was entirely due to the poor
performance of ADMARC. IBs other than ADMARC matched the overall market performance
in 1999 and far outperformed others in 2000. This general performance is surprising given that
the general assumption has been that IBs primarily purchase the crop of those very poor
smallholders who do not have very effective clubs and who have less access to both credit and
infrastructure. One would have expected IB sales to reflect a rather poorer mix of crop,
especially if one adds that some of the IBs lacked tobacco grading experience.

         As regards credit, there is no doubt that some IB purchases contributed to loan default as
growers deliberately sold all or part of their tobacco through side channels (including the IBs and
estates) in order to bypass their stop orders on the auction floors. However, the two seasons of
most prominent IB purchases--1997 and 1998--were actually the two seasons when MRFC
recorded its highest levels of loan recoveries �90%. In the last three years of more limited IB
operations, MRFC loan recoveries have been substantially lower--at just under 80%. Hence, on
the big picture, there is no correlation between IB operations and credit default, despite the
periodic incidence of side-selling by willful defaulters. Even if licensed IBs never existed, the
side-selling would probably have taken place anyway through estates, and indeed, most of the
side-selling which has taken place has continued to occur in this manner.

         The rising incidence of tobacco theft was probably associated with the increase in
licensed IB participation. The most common form of this practice was for estate tenants and
managers to sell the estate's tobacco off the farm and thereby deprive the estate's owner of the
tobacco and the revenue stream thereof. This seems to have occurred on a substantial scale in the
case of absentee owned estates, but also adversely affected owner-managed estates, including
some of the better managed estates in the country. However, opportunism on the part of estate
workers and managers didn't begin with the introduction of the IB system and won't end as long
as returns to workers remain very low.

         Hence, the IB system accelerated the decline in the burley tenancy, although weaknesses
in that system were evident for a number of years prior to the introduction of IBs. The rise of
non-legitimate sales of tobacco in rural areas occurred within a general climate of increased
crime and theft in those areas during the mid-to-late 1990s. The IBs didn't cause this. Many of
the IBs came from the same pool of people �estate owners--who were probably most adversely
affected by the illegitimate sales. Given the difficult financial situation facing many estates, there
is little reason to believe that their IB operations contributed either more or less to the incidence
of theft than the IB operations of any other class of IBs.

         The IB system was wound down in 2000, precisely at the wrong time. By that year, most
of the excesses and abuses of the system had been eliminated and a more stable structure of a



                                                  43

few hundred more experienced IBs had emerged. Here and there people were still using their IB
license to by-pass this or that responsibility, but, for the most part, the vast majority of fly-by-
night operators had burnt their fingers and exited from the IB function. By 2000, the IB function
was beginning to evolve with a sub-set of IB operators providing inputs on credit and selected
other services to growers. That evolution ceased with the official announcement that the IB
system would end.

        In September 2000, the Cabinet Committee on the Economy directed the Ministry of
Agriculture and Irrigation to introduce a new arrangement called the Designated Tobacco Buyers
Scheme (DTBS).48 This would be a modification of the IB program in that a more limited set of
intermediate buyers would be licensed and they would be permitted to operate only at designated
sites. Only persons or organizations with tobacco handling facilities--i.e. a baling press and
certified scales--would be eligible to participate. Certain locations would be designated as
tobacco selling sites, presumably in close vicinity to a local government (agricultural extension)
office. A series of rules and regulations on permissible transactions would be devised and
enforced by local DTBS Committees, comprised primarily of TAMA and government
representatives.

        The DTBS was designed to reduce some of the opportunistic behavior associated with the
previous un-regulated IB system and to bring more transparency to the relationship between
growers and intermediate buyers. There is some scope that the designated buyer sites could
evolve into small marketplaces, where, provided there is a competition among multiple buyers, a
mini-auction could develop. However, how the DTBS will develop would seem to depend
primarily on localized power structures and whether the local `committees' determine that the
system should best serve growers rather than basically serve themselves and their designated
buyer colleagues. At present, neither the Tobacco Control Commission nor the Ministry of
Agriculture and Irrigation have the capacity to effectively monitor and police the functioning of
these designated buyer mini-markets.




48This was in response to a cabinet paper on the subject submitted by the Ministry in January 2000 and then
resubmitted in July 2000 after revisions were made. The uncertainty regarding the status of the IB program in the
months leading up to the marketing season in 2000 contributed to the reduction of licensed IBs that season.



                                                         44

                           Table 10: Intermediate Buyers: Performance on Burley Tobacco Sales, 1994 � 2001

      Year          # of Licensed Ibs     Quantity Sold    Share of National   Average Sales      Average National   IB Ave. Price/
                                            (Tons)             Crop (%)      Price for Ibs (US $)   Price (US$)    National Ave. Price
                                                                                                                         (%)

IBs excluding ADMARC
1994                       79                 398                0.6                1.28               1.29        99.7
1995                      1074               4854                4.8                1.42               1.48        95.5
1996                      3106               3834                3.3                1.59               1.61        98.5
1997                      4012              19287                14.4               1.51               1.53        98.7
1998                      3239              13889                12.2               1.27               1.30        98.2
1999                      1402               3089                2.8                1.38               1.38        100
2000                       671               3547                2.5                1.30               1.02        127.8
2001                       148                542                0.5                1.04               1.10        94.8

IBs including ADMARC
1999                      1405               8909                8.0                1.28               1.38        92.4
2000                       674               8239                5.8                0.91               1.02        89.5
2001                       154               1177                1.0                1.05               1.10        95.2

ADMARC IB Operation
1999                        3                5820                5.2                1.22               1.38        88.2
2000                        3                4692                3.3                0.62               1.02        60.6
2001                        6                 635                0.6                1.05               1.10        95.6

Source: Tobacco Control Commission




                                                                 45

         It has now been nearly two years since the Cabinet Directive on this issue, yet the DTBS
is not yet in place49, ostensibly because there is a lack of funds to undertake the set of planning
meetings needed to elaborate the detailed arrangements for the scheme and put in place the local
committees which will implement the program. These planning costs are not very high50 and the
failure to mobilize even the limited funds to set up the scheme does not inspire confidence about
near term capacities to provide proper oversight and evaluation of the performance of the
scheme. Once again, inertia prevails.

         The current high costs of market intermediation and the failure of the existing system to
give rise to innovative solutions to the logistics, cash flow, and other problems faced by the
majority of smallholder tobacco growers indicate that `business as usual' is not a viable strategy
for the industry and that the on-going state of policy paralysis must end. There are no easy, quick
fix solutions and the industry is not in a position to effectively negotiate a path of radical change
in the system of marketing and supply chain management for the tobacco crop. However, it is
imperative that a process be initiated which will yield the types of innovations that will radically
reduce the marketing and transaction costs faced by Malawi's smallholder growers.

         For this to happen, the Malawi government will need to act. Its first set of actions could
include:

    � Issuing an directive to AHL to reduce its auction fee and stop-order charge and to
         otherwise consider AHL as a regulated monopoly;
    � Applying pressure--via the large shareholding by ADMARC in AHL-- in order to
         increase the representation of smallholder farmers on the board of AHL
    � Issuing a directive for a time-bound re-assessment of the existing level of statutory levies
         being applied with a view to reducing these51;
    � Directing the responsible agencies to revamp the tobacco grower registration system and
         to develop a modality by which the withholding (income) tax is assigned only to those
         estate growers who meet the income thresholds

The next set of actions require choices among various options. The responsibilities for the
satellite depot system most likely need to be re-assigned. Government could either:

    � Encourage AHL to take over this function and help negotiate a reasonable fee structure
         for this work; or
    � Put the management of a national or regional/district system of satellite depots up for
         competitive tender, select suitable managers, and oversee their performance

The advantage of either of the above options over the current system is the potential for
integrating into the satellite depot system a partial payment arrangement whereby farmers might


49This meant that in 2001 and 2002 there was a significant sale of smallholder tobacco to estates who acted as
intermediate buyers, yet unlicensed ones.
50Estimated at approximately K500,000 or just under $7,000 at the current exchange rate.
51A possible exception might be the ARET levy, depending upon there being in place an agreed work program for
research and advisory services. Initial proposals tabled in January 2003 to modify the levies would result in virtually
no improvement in smallholder farmer incomes.



                                                          46

get paid 15-25% of the estimated value of their crop upon delivery to the depot. This will enable
them to finance their most pressing cash needs (i.e. for funeral expenses, basic necessitates). This
arrangement would reduce the need for poor farmers to sell tobacco through an intermediate
buyer, the primary advantage of which is the immediate cash payment. This partial payment
could be linked into the AHL's grower database and payment system. Putting the satellite depot
system under more professional management also provides the scope for the future evolution of
these depots, perhaps to become off-farm tobacco curing sites, thereby economizing on wood
pole and thatch requirements for individual farmers52 and possibly also cutting back on the theft
of tobacco in rural areas.53

         There are other options available to government to stimulate competition and innovation
in the existing tobacco supply chain. These include at least the following:

    � Breaking up the monopoly structure in the auctioning function, perhaps by having AHL
         sell one or two of its auction floors to other companies, retaining the remainder for its
         own operations. This would stimulate competition among different auction floors,
         improving services and likely resulting in one or more auction companies moving to
         decentralize their operations, either to the level of satellite depots or by setting up some
         smaller auctions/market-places in their primary region; or

    � Simply provide direct encouragement and support for the emergence of several smaller
         auction/marketplace facilities in areas such as Kasungu, Machinga, and other important
         production locales which are relatively distant from the existing auction floors.

         Another option concerns permitting the sale of Malawi tobacco outside of the auctions.
Malawi is one of the few major tobacco producing countries in which the bulk of sales are
transacted through auctions. Only Zimbabwe and Canada rely primarily on auctions, while for
India a small proportion of the crop is marketed in this way. Until two years ago, tobacco sales in
the United States were almost entirely via auction. However, direct contracting was piloted in
2000 and by the subsequent year some 70% of the US tobacco crop was sold under contract
between growers and cigarette manufacturers.

         According to the USDA's Agricultural Marketing Service, in 1998, an estimated 80% of
world flue-cured production and 50% of world burley production were marketed under contract,
either through private companies or state-owned enterprises. With the more recent shift in the
sales method for US tobacco, it is likely that the proportions of world flue-cured and burley
tobacco production now marketed under contract exceeds 90% and 65%, respectively. Table 11
provides a summary of the primary modes for tobacco sales among major producing countries.




52More permanent structures, including the use of plastics and other materials in place of wood poles, might be
developed.
53A third option is to facilitate the development of a `designated service provider' system whereby existing
transporters, rural retail operators, and perhaps others would be supported in developing a diversified and integrated
set of business services �including providing the functions of a satellite depot.



                                                          47

              Table 11: Tobacco Marketing Arrangements for Leading World Producers

                              Contract        State Enterprise      Auction       Market
                              Farming           Procurement                      Purchase
   China                                      Monopsony
   India                  Significant                           For Export    Significant
                                                                Varieties
   Brazil                 Primary                                             Residual
   United States          Primary                               Residual
   Turkey                                     Significant                     Significant
   Zimbabwe                                                     Entire Crop
   Malawi                                                       Entire Crop
   Mexico                 Primary                                             Residual
   Argentina              Primary                                             Residual
   Korea                                      Monopsony
   Thailand               Residual            Up to Quota                     Residual
   Italy                                      Monopsony

        The modes of contracting differ among countries.54 In the United States, the initial
arrangements essentially are marketing contracts, involving a pricing and delivery schedule and a
network of crop receiving and inspection stations set up by the buyers. The contracts do,
however, make reference to certain production practices and other issues related to quality, use
of only approved chemicals, grading procedures, etc.

        In much of Latin America, the normal arrangement combines a marketing contract with
the provision (on credit) of production inputs and also the provision of extension services by the
buying companies. Producer and industry representatives typically negotiate a series of prices for
various grades, taking into account the costs of production and the overall market situation. The
contracts provided an assured market outlet for growers and typically entail some form of risk
management arrangement in the event of crop loss (say, due to drought).

        Access to inputs, production information, financing, and certain management skills are all
limiting factors in Malawi. The capacity of its smallholder farmers to manage risks is also a
major constraint. Some such elements could be incorporated into production/marketing contracts
within Malawi, provided that an institutional framework were put in place to enable the buyers to
effectively implement and enforce these contracts. Savings would also be achieved by growers
on logistics and transaction costs since the buyers would likely arrange transport and as there
would likely be a collective bargaining arrangement to work out standardized contracts and a
price schedule.

        In other contract farming contexts there are typically concerns about the unequal
bargaining power between growers and the normally large agribusiness firms with whom they
contract. Such unequal bargaining power can translate itself into unfavorable terms and
conditions for the growers and the concentration of benefits--in terms of improved quality


54See, for example, Snell and Green (2000); and Tiller (2001)



                                                        48

control, supply management, etc.--to the buyers. In the Malawi context this unequal bargaining
position certainly exists in an extreme proportion--to the point where the annual turnover of one
of the parent companies of the major buyers is larger than the GDP of Malawi--yet this remains
the case whether sales are conducted through the auctions or through direct contracting. It should
therefore not be a factor in weighing this policy option.

        It is not obvious that direct contracting is the most suitable arrangement for the Malawi
tobacco industry as a whole given the fragmented structure of production. The three large
international buyers are decidedly unenthusiastic about this option. While this is their normal
mode of purchase in such countries as Brazil, Argentina, and Mexico, the companies have
experienced massive problems of `side-selling' in Tanzania and are deeply concerned about the
breakdown of credit repayment discipline in Malawi. They perceive that the administrative costs
of contracting large numbers of Malawian growers would be huge.

        In this context, one might assume that the companies would `cherry pick', initially
entering into contracts with some larger growers and perhaps with parts of NASFAM. This
would serve as their trial for direct contracting in Malawi. Already for the 2002/03 season the
international buyers sourced finance for several larger growers of flue-cured tobacco, although
ostensibly this tobacco was sold on the auctions rather than directly to the buyers. Among
consenting adults, a direct grower �buyer relationship should be permitted.              55


        The most likely (as well as the most favorable) scenario would feature a combination of
selected contracting and the continuation of (an improved version of) the auction system. Large
numbers of Malawian smallholders will continue to need to sell over the auction floors. And, if
marketing and transaction costs are reduced, many estates will want to continue to sell over the
auction floors, especially if they perceive that they can obtain higher prices than the prevailing
contracted prices in any particular year. Some growers may choose to sell under both systems to
take advantage of certain benefits of each. Selected direct contracting could contribute to the
reduction of the congestion problem at the Kanengo floors, provided that the contracts specify
delivery times to the buyers either before or after the main bottleneck period. Simply the threat
of direct contracting will force AHL to act to reduce the costs and other problems associated with
the auction-linked supply chain.

        It is recommended that existing regulations be changed to permit the direct
contracting of tobacco production. If there is any reason to believe that this might cause a
major disruption in the marketing system, then steps should be taken to phase in this regulatory
change, perhaps permitting direct contracting for flue-cured production in the 2003/2004 season
and perhaps also for burley production within a specified pilot set of locations. Based on the
monitored experiences, the more general permission for direct contracting could be put in place.
This phased introduction of direct contracting will also provide some time to put in place other
measures--such as the revamped grower registration system and a credit reporting system--
which will help reduce breach of contractual terms.


55A draft Cabinet Paper in January 2003 proposed to ban contract farming in the industry, ostensibly to prevent
farmers from being `exploited'. There is no reason why farmers would be any more `exploited' under a contractual
arrangement than they are now and very likely these farmers would have much greater predictability about their
seasonal incomes.



                                                       49

        The overall scope for cost reductions in the tobacco supply chain are significant and these
can be achieved in a short period of time. At a minimum, one could achieve the following cost
savings associated with the tobacco of smallholder farmers:

        Source                                                        Savings
                                                               U.S. Cents/Kg. of Tobacco

        Lower int'l transport costs via Nacala                        8.0
        Remove withholding tax for clubs                              7.0
        More competitive transport rates internally                   3.0
        Reduction of AHL auction fee                                  1.5
        Reduction of AHL stop-order fee                               1.5
        Reduction of TAMA classification levy                         0.25
        Reduction of hessian levy                                     0.25
        Total                                                         21.5



Raising Productivity


        While reducing marketing and transaction costs in the tobacco supply chain should be the
central focus on policy-makers, the benefits which will accrue from this will not be sustained
unless, over the medium term, the industry players can reverse the downward trend in tobacco
yields. As noted earlier, low and declining tobacco yields in Malawi are a result of an array of
factors, including soil infertility, low and frequently incorrect application of fertilizer, the build
up of pests and diseases due to improper crop rotations, the use of recycled and low quality seed,
poor construction/maintenance of tobacco curing barns, limitations of technical knowledge about
the cultivation and curing of the crop, and labor-related problems--such as inappropriate timing
of transplantation, insufficient weeding, delayed harvesting.

        A multi-pronged approach is needed to address these problems and constraints, some of
which (i.e. soil infertility) extend well beyond the issue of tobacco crop management itself. In the
past, government and some other stakeholders have placed predominant emphasis on the
problem of access to fertilizer, or more specifically, on the provision of credit for farmers to
procure fertilizer. This is undoubtedly an important issue, yet the improvement of cultivation
practices for the tobacco crop and developing more sustainable approaches to the curing of the
tobacco leaf are of equal importance, yet have been given inadequate attention.

        ARET is in the process of developing plans for a "Back to Basics" program of
communications and advisory services, with an increased orientation to smallholder farmers. The
organization lacks the necessary manpower to reach large numbers of smallholder growers and
thus close collaboration with the extension staff of the Ministry of Agriculture and Irrigation will
be necessary. There may also be the need to revamp part of ARET's research program to put
more emphasis on financial and economic constraints and to investigate improved cultural
practices under typical smallholder farm conditions.



                                                  50

        As regards credit, the challenge is to tackle the fundamental institutional infrastructure
for a viable system, rather than simply develop new schemes from year-to-year to throw
resources at past loan defaulters in the hope that this generates an additional crop and that some
of that money is repaid. A credit repayment discipline needs to be re-established, yet this is
precisely impossible in the face of new schemes popping up every year and enabling past
defaulters to access new loans on the basis of clever misrepresentation of their past records or
even of who they are.

        A proper institutional framework needs to be (re-) established involving a national
identity card, a credit reporting system, a re-vamped grower registration system, and perhaps,
special courts to swiftly handle the loan recovery cases of financial institutions. When the basic
foundations are put in place, financial institutions will be more able and willing to extend credit
to the agricultural sector and agribusiness companies will do so as well. Without such
foundations, the risks of lending are simply too high and the sector will simply bounce from one
scheme to the next.

                                        CONCLUSIONS

        Malawi is one of the world's largest exporters of tobacco. Despite global and national
efforts to control tobacco use because of its highly detrimental health effects, world demand for
tobacco will likely continue to grow at least over the next decade and probably well beyond that.
Ample market opportunities should be available to Malawi's industry. However, with the general
shift in international demand from higher income to low and middle income countries, there is an
expectation of some continued downward pressures on international tobacco prices. Hence, the
challenge for Malawi is to raise its productivity and reduce its production and marketing costs so
that it can remain competitive and generate a stream of profits even under a scenario of lower
export prices.

        Malawi's success in meeting this challenge is crucial to the overall development of its
economy and to its ability to reduce poverty. Tobacco is Malawi's dominant industry and will
remain so at least in the medium term (i.e. 5-7 years) even if concerted progress is made in
attracting investment and diversifying the economy. Even this diversification will require, in
part, the revitalization of the profitability of tobacco cultivation, given the wide participation in
this activity and its scope for generating demand for other goods and services as well as
investable savings.

        This paper has highlighted some of the recent gains made in the Malawi tobacco
subsector, yet also emphasized its precarious state as a result of declining productivity and
profitability, widespread in-fighting among stakeholders, and a situation of policy paralysis.
Many of the constraints and inefficiencies within the subsector derive from vested interests and
from a lack of transparency and accountability regarding actions and performance. The policy
paralysis needs to be broken and concerted action taken which results in more money going into
the pockets of the country's tobacco growers. This paper has laid out an agenda or set of options
for the main stakeholders--including the Government of Malawi--to pursue.




                                                   51

                                     REFERENCES

Banda, Albert, Charles Mataya, Flora Munthali, Ron Ngwira, and Gavin Olney (1998) Towards
       Increased Productivity and Sustainability in Smallholder Burley Tobacco. Study for the
       World Bank, contracted to Agricultural Cooperative Development International.

Gossage, Steven and Jaspar Steele. (1997 Estate Land Utilization Study. Lilongwe.

Hirschman, Albert (1970) Exit, Voice, and Loyalty: Responses to Decline in Firms,
       Organizations, and States. Cambridge: Harvard University Press.

Jaffee, Steven (1997) Malawi Agriculture: Recent Structural Transformation and Future
       Prospects. World Bank.

Jaffee, Steven, Richard Mkandawire, and Sandra Bertoli (1991) Migrant Smallholders: Tenant
       and Laborer Particiation, Remuneration, and Social Welfare Within Malawi's Expanding
       Estate Sub-sector. Study for USAID. Binghamton, NY: Institute for Development
       Anthropology.

Jungbluth, Frauke and Stanley Hiwa (1999) Malawi: Fertilizer Sector Issues and Policy Options
       Paper. World Bank.

Kasnakoglu, Haluk and Erol Cakmak (2000) Economics of Tobacco in Turkey. Prepared for the
       FAO.

Keyser, John and Victor Lungu (1997) Malawi: Agricultural Comparative Advantage. Paper
       prepared for the World Bank.

Kille, Turville, George Olesh, and Richard Buckingham (1998) Malawi Tobacco Export and
       Marketing Study. Prepared for the Government of the Republic of Malawi. Cargill
       Technical Services.

Kydd, Jonathan and Robert Christiansen (1982) "Structural Change in Malawi since
       Independence: Consequences of a Development Strategy Based on Large-Scale
       Agriculture," World Development, Vol. 10, No. 5, pp. 355-75.

Mergos, G. (2001) Tobacco: Supply, Demand, and Trade Projections, 2005-2010. Prepared for
       the FAO.

Mkandawire, Richard, Steven Jaffee, and Sandra Bertoli (1990) Beyond Dualism: The Changing
       Face of the Leasehold Estate Subsector in Malawi. Study for USAID. Binghamton, N.Y.:
       Institute for Development Anthropology.




                                              52

Olney, Gaving, Charles Chilimampunga, Peter Hopcraft, Aggrey Kawonga, and Francis
        Majankono (1998) Managing the Emerging Large Farm Crisis and Options for Land
        Resettlement. Paper prepared for the Ministry of Lands, Housing, Physical Planning &
        Surveys, and for the World Bank.

Snell, William and Daniel Green (2000) "Tobacco Contracting Issues and Update for 2001" in
        Issues Surrounding the Debate on Direct Marketing of Tobacco, University of Kentucky.

Tiller, Kelly (2001) "Tobacco Issues: Contracting and Use of Tobacco Settlement Payments,"
        Agricultural Outlook Forum. Washington: USDA.

United States Department of Agriculture, Agricultural Attache Reports, various.

Westlake, Michael (1999) Malawi: Structure of Fertilizer and Transport Costs. Prepared for the
        European Commission.

Wilshaw, Colin (1994) A Century of Growth: Malawi's Tobacco Industry, 1893-1993. Blantyre:
        Central Africana Ltd.

Zeller, Manfred (1997) Growth Linkages of Smallholder Tobacco Production in Malawi. Paper
        prepared for the World Bank.




                                              53

       Annex: 1: Quality Results of TAMA Tobacco Classification System Results
                           for Burley Tobacco, 1992 to 2001

                    Proportion (%) of the Crop Classified by Grade
Year             1st                2nd               3rd             4th/5th
1992             1.4                20.2              47.9            30.5
1993             0.6                15.9              45.1            38.5
1994             3.6                20.1              50.1            26.2
1995             4.6                24.2              51.6            19.6
1996             3.3                21.1              54.9            20.7
1997             3.6                24.6              53.2            18.7
1998             4.3                22.8              52.2            20.8
1999             3.7                24.1              53.0            19.2
2000             2.5                19.0              56.8            21.8
2001             3.6                22.8              52.7            20.9

        Proportion (%) of the Crop Classified as Either Mouldy or Mixed Grade
Year                                   Mouldy                     Mixed Grade*
1992                                     19.7                         10.7
1993                                     26.2                         13.2
1994                                     14.6                         11.3
1995                                     13.8                         10.7
1996                                     20.5                         7.6
1997                                     11.8                         7.9
1998                                      9.6                         9.0
1999                                     10.2                         8.6
2000                                      9.7                         13.8
2001                                      3.0                         13.3

*Requiring re-grading and re-baling.




                                              54

                                Africa Region Working Paper Series
 Series #                           Title                               Date           Author

ARWPS 1   Progress in Public Expenditure Management in Africa:   January 1999    C. Kostopoulos
          Evidence from World Bank Surveys


ARWPS 2   Toward Inclusive and Sustainable Development in the    March 1999      Markus Kostner
          Democratic Republic of the Congo

ARWPS 3   Business Taxation in a Low-Revenue Economy: A          June 1999       Ritva Reinikka
          Study on Uganda in Comparison with Neighboring                         Duanjie Chen
          Countries

ARWPS 4   Pensions and Social Security in Sub-Saharan Africa:    October 1999    Luca Barbone
          Issues and Options                                                     Luis-A. Sanchez B.

ARWPS 5   Forest Taxes, Government Revenues and the              January 2000    Luca Barbone
          Sustainable Exploitation of Tropical Forests                           Juan Zalduendo

ARWPS 6   The Cost of Doing Business: Firms' Experience with     June 2000       Jacob Svensson
          Corruption in Uganda

ARWPS 7   On the Recent Trade Performance of Sub-Saharan         August 2000     Francis Ng and
          African Countries: Cause for Hope or More of the Same                  Alexander J. Yeats

ARWPS 8   Foreign Direct Investment in Africa: Old Tales and New November 2000   Miria Pigato
          Evidence

ARWPS 9   The Macro Implications of HIV/AIDS in South Africa:    November 2000   Channing Arndt
          A Preliminary Assessment                                               Jeffrey D. Lewis

ARWPS 10  Revisiting Growth and Convergence: Is Africa Catching  December 2000   C. G. Tsangarides
          Up?

ARWPS 11  Spending on Safety Nets for the Poor: How Much, for    January 2001    William J. Smith
          How Many? The Case of Malawi

ARWPS 12  Tourism in Africa                                      February 2001   Iain T. Christie
                                                                                 D. E. Crompton

ARWPS 13  Conflict Diamonds                                      February 2001   Louis Goreux

ARWPS 14  Reform and Opportunity: The Changing Role and          March 2001      Jeffrey D. Lewis
          Patterns of Trade in South Africa and SADC

ARWPS 15  The Foreign Direct Investment Environment in Africa    March 2001      Miria Pigato

ARWPS 16  Choice of Exchange Rate Regimes for Developing         April 2001      Fahrettin Yagci
          Countries

ARWPS 17  Export Processing Zones: Has Africa Missed the Boat?   May 2001        Peter L. Watson
          Not yet!

ARWPS 18  Rural Infrastructure in Africa: Policy Directions      June 2001       Robert Fishbein


                                                    55

                               Africa Region Working Paper Series
 Series #                          Title                               Date           Author
ARWPS 19  Changes in Poverty in Madagascar: 1993-1999           July 2001       S. Paternostro
                                                                                J. Razafindravonona
                                                                                David Stifel

ARWPS 20  Information and Communication Technology, Poverty,    August 2001     Miria Pigato
          and Development in sub-Saharan Africa and South Asia

ARWPS 21  Handling Hierarchy in Decentralized Settings:         September 2001  Navin Girishankar
          Governance Underpinnings of School Performance in                     A. Alemayehu
          Tikur Inchini, West Shewa Zone, Oromia Region                         Yusuf Ahmad

ARWPS 22  Child Malnutrition in Ethiopia: Can Maternal          October 2001    Luc Christiaensen
          Knowledge Augment The Role of Income?                                 Harold Alderman

ARWPS 23  Child Soldiers: Preventing,Demobilizing and           November 2001   Beth Verhey
          Reintegrating

ARWPS 24  The Budget and Medium-Term Expenditure Framework      December 2001   David L. Bevan
          in Uganda

ARWPS 25  Design and Implementation of Financial Management     January 2002    Guenter Heidenhof
          Systems: An African Perspective                                       H. Grandvoinnet
                                                                                Daryoush Kianpour
                                                                                B. Rezaian

ARWPS 26  What Can Africa Expect From Its Traditional Exports?  February 2002   Francis Ng
                                                                                Alexander Yeats

ARWPS 27  Free Trade Agreements and the SADC Economies          February 2002   Jeffrey D. Lewis
                                                                                Sherman Robinson
                                                                                Karen Thierfelder

ARWPS 28  Medium Term Expenditure Frameworks: From Concept      February 2002   P. Le Houerou
          to Practice. Preliminary Lessons from Africa                          Robert Taliercio

ARWPS 29  The Changing Distribution of Public Education         February 2002   Samer Al-Samarrai
          Expenditure in Malawi                                                 Hassan Zaman

ARWPS 30  Post-Conflict Recovery in Africa: An Agenda for the   April 2002      Serge Michailof
          Africa Region                                                         Markus Kostner
                                                                                Xavier Devictor

ARWPS 31  Efficiency of Public Expenditure Distribution and     May 2002        Xiao Ye
          Beyond: A report on Ghana's 2000 Public Expenditure                   S. Canagaraja
          Tracking Survey in the Sectors of Primary Health and
          Education

ARWPS 32  Promoting Growth and Employment in South Africa       June 2002       Jeffrey D.Lewis

ARWPS 33  Addressing Gender Issues in Demobilization and        August 2002     N. de Watteville
          Reintegration Programs



                                                   56

                               Africa Region Working Paper Series
 Series #                           Title                                Date            Author

ARWPS 34  Putting Welfare on the Map in Madagascar                August 2002     Johan A. Mistiaen
                                                                                  Berk Soler
                                                                                  T. Razafimanantena
                                                                                  J. Razafindravonona

ARWPS 35  A Review of the Rural Firewood Market Strategy in       August 2002     Gerald Foley
          West Africa                                                             Paul Kerkhof
                                                                                  Djibrilla
                                                                                  Madougou

ARWPS 36  Patterns of Governance in Africa                        September 2002  Brian D. Levy

ARWPS 37  Obstacles and Opportunities for Senegal's International September 2002  Stephen Golub
          Competitiveness: Case Studies of the Peanut Oil,                        Ahmadou Aly
          Fishing and Textile Industries                                          Mbaye

ARWPS 38  A Macroeconomic Framework for Poverty Reduction         October 2002    S. Devarajan
          Strategy Papers : With an Application to Zambia                         Delfin S. Go

ARWPS 39  The Impact of Cash Budgets on Poverty Reduction in      November 2002   Hinh T. Dinh
          Zambia: A Case Study of the Conflict between Well                       Abebe Adugna
          Intentioned Macroeconomic Policy and Service                            Bernard Myers
          Delivery to the Poor

ARWPS 40  Decentralization in Africa: A Stocktaking Survey        November 2002   Stephen N.
                                                                                  Ndegwa

ARWPS 41  An Industry Level Analysis of Manufacturing             December 2002   Professor A.
          Productivity in Senegal                                                 Mbaye

ARWPS 42  Tanzania's Cotton Sector: Constraints and Challenges    December 2002   John Baffes
          in a Global Environment

ARWPS 43  Analyzing Financial and Private Sector Linkages in      January 2003    Abayomi Alawode
          Africa

ARWPS 44  Modernizing Africa's Agro-Food System: Analytical       February 2003   Steven Jaffee
          Framework and Implications for Operations                               Ron Kopicki
                                                                                  Patrick Labaste
                                                                                  Iain Christie

ARWPS 45  Public Expenditure Performance in Rwanda                March 2003      Hippolyte Fofack
                                                                                  C. Obidegwu
                                                                                  Robert Ngong

ARWPS 46  Senegal Tourism Sector Study                            March 2003      Elizabeth Crompton
                                                                                  Iain T. Christie


ARWPS 47  Reforming the Cotton Sector in SSA                      March 2003      Louis Goreux
                                                                                  John Macrae


ARWPS 48  HIV/AIDS, Human Capital, and Economic Growth            April 2003      Channing Arndt
          Prospects for Mozambique
                                                   57

                               Africa Region Working Paper Series
 Series #                          Title                              Date          Author
          Prospects for Mozambique

ARWPS 49  Rural and Micro Finance Regulation in Ghana:          June 2003    William F. Steel
          Implications for Development and Performance of the                David O. Andah
          Industry

ARWPS 50  Microfinance Regulation in Benin: Implications of the June 2003    K. Ouattara
          PARMEC LAW for Development and Performance of
          the Industry

ARWPS 51  Microfinance Regulation in Tanzania: Implications for June 2003    Bikki Randhawa
          Development and Performance of the Industry                        Joselito Gallardo

ARWPS 52  Regional Integration in Central Africa: Key Issues    June 2003    Ali Zafar
                                                                             Keiko Kubota

ARWPS 53  Evaluating Banking Supervision in Africa              June 2003    Abayomi Alawode


ARWPS 54  Microfinance Institutions' Response in Conflict       June 2003    Marilyn S. Manalo
          Environments : Eritrea- Savings and Micro Credit
          Program; West Bank and Gaza � Palestine for Credit
          and Development; Haiti � Micro Credit National, S.A.

AWPS 55   Malawi's Tobacco Sector: Standing on One Strong Leg   June 2003    Steven Jaffee
          is Better than on None




                                                  58