Developing futures and swap markets for tea: opportunities,

risks and conditions for success

Report to be presented at the 23rd session of the FAO

Intergovernmental Group on Tea, May 2018
(A Subsidiary Body of the FAO Committee on Commodity Problems)

DRAFT

1
 Contents

Executive summary

Introduction

1. Market structures and marketing arrangements in the main producing countries

2. Tea prices and volatility

3. How well are the prices for different varieties, grades and markets connected?

4. Conditions for the successful creation of tea futures and swap contracts

5. Creating tea futures and swap markets: potential challenges, and possible ways to address them

6. Building inclusive markets: how to bring growers on board

Concluding remarks

References

2
EXECUTIVE SUMMARY
Tea growers, factories, traders, exporters and blenders are exposed to considerable tea price risk,
endangering their sustainability. The FAO Intergovernmental Group on Tea (FAO – IGG/Tea), as well as
industry bodies and governments in several producing countries, have identified tea futures as a way to
empower the tea industry to deal with this risk. More recently, tea swaps have been suggested as a risk
mitigation tool.

A tea futures exchange will not only provide a platform for risk management, it will also improve price
discovery in the tea market, which in particular favors the weaker parties in the tea value chain. Beyond
their role of giving farmers more predictable revenues and traders and processors better manageable
profit margins, tea futures have the potential to unlock finance for growers, so that they can better
invest in future crop production (this is because financiers can be more certain of the growers’ ability to
repay if these growers manage their price risks).
While futures contracts for coffee have been actively traded since 1882 and contracts for cocoa since
1925, there are no futures contracts for tea. Efforts to create one so far been unsuccessful. Some in the
tea industry believe that with its more than 3,000 varieties, tea is too heterogeneous to trade on a
futures exchange. Tea is heterogeneous both over season and region, and even intra-region depending
on the stock from which harvest was made and the periodicity of the harvest. Thus, it is difficult to agree
on a “reference variety” that can act as a standard. Others agree that it will be difficult and complex –
one advocate of futures markets in India, for example, suggested a futures contract might require to
include as valid for delivery as many as 50 different grades of tea – but feel it is feasible and worthwhile.

The tea sector is large. However, analysis shows that in fact it consists of a number of poorly-connected
segments, the main ones being China for green tea; Colombo for orthodox black tea; and separate CTC
black tea markets in North India, South India and Kenya. These markets range in size from 1 to 4 billion
US$. Price correlations between these markets are low. The separate behavior of the three CTC black
tea markets may be the result of a lack of a central reference market – thus, once there is a futures
contract, prices may become better integrated.

Of the various segments of the tea market, China’s green tea segment is the largest, worth around US$ 4
billion. The futures contract would have to be focused on the domestic market, rather than on exports,
but the market size is more than enough to support a viable futures contract. However, year-to-year
price changes, are low. Intra-year prices are not readily available for a longer time period, but the 2017
indices for black and green tea suggest the lowest prices received by tea factories in that year were
about two thirds of the highest prices – a level of price uncertainty that is sufficient to warrant an
interest in hedging. Nevertheless, in the absence of auctions or other organized market places, the
market appears too fragmented to readily organize a futures market. Before a futures contract becomes
feasible, an effort needs to be made to better organize the physical market, e.g., by building on recent
efforts to gather and publish price data, and to agree on common quality standards.

Global black tea markets are well organized, with auctions not only acting as nodal points for physical
trade flows, but also offering a mechanism through which market participants agree on grades and
standards. A futures exchange can build on this, with physical deliveries organized around the auction

3
infrastructure, or using auction prices to establish a generic reference price for tea. Unfortunately, black
tea markets are segmented. The CTC tea market in North India is the largest, and a futures contract
could be organized around the tea trade that flows through its auctions.

For other market segments, it would be more difficult. South Indian CTC tea futures, if they share a
platform with North Indian CTC tea, can perhaps feed on the latter’s liquidity. For the smaller,
segmented markets in Sri Lanka and Kenya (US$ 1.2 to US$ 1.4 billion of underlying value, compared to a
US$ 2 billion underlying market in North India) it will be more difficult. The costs of introducing tea
futures would be high because there are no established futures exchanges that could add tea to their
product offerings. To be able to introduce a futures contract without having to cover the fixed costs of
creating an exchange, and to gain immediate access to a large pool of speculators, the tea sector in
these countries could collaborate with an overseas futures exchange to offer tea futures as a cloud
solution.

Given the complexity of the sector, the initiative to introduce tea futures has to come from the tea
sector itself. They would have to engage established futures markets to obtain relevant expertise, gain
access to trading and clearing platforms etc., but without such a drive by physical market players the
futures industry is not likely to move ahead. The industry has to chose the right organizational model –
leverage existing technologies and platforms, rather than invest in their own; help design contract
specifications that ensure that the futures contract will reflect physical market conditions, either
through a sound physical delivery system or through an index contract; and support the initial efforts of
an exchange to build liquidity in a new tea futures contract.
In that process, the industry has to deal with its own doubts about designing a tea futures contract that
adequately represents the realities of tea trade. It is worthwhile to consider the experience of the coffee
sector, which deals with the wide variety of coffees traded by having three levels of discounts and
premiums:
- For specific grades (origin countries), set from time to time;
- For specific delivery locations, set from time to time;
- And for specific deliveries, set by a group of exchange experts who will actually taste the coffee
that is delivered to determine whether it merits an additional premium or discount.
Thus, all coffee delivered to an exchange is valued very closely to its actual value on the market. Or in
other words, the coffee market shows that a standardized futures contract does not mean that physical
coffee gets treated in a standardized manner – rather, its individual characteristics remain recognized
and rewarded. It is difficult to understand why a similar mechanism would not be feasible for a tea
futures contract.

If a tea futures contract can be created, it is not automatically inclusive. The deliveries would have to be
at an ex-factory level, which excludes most growers from actually deliver against a futures contract. But
in the experience of other commodities, typically growers do not use futures markets directly (although
this may now change, as a result of the spread of mobile phones and electronic money). Rather, growers
have been benefiting from futures markets because these provide them with superior market

4
information; and they can indirectly use futures exchanges through cooperatives, processors and
traders. Donor agencies and governments may wish to give technical assistance to this effect.

Tea swaps have been mentioned as an alternative to futures contracts. But while these do provide a way
to manage price risks, but they cannot replace and organized futures exchange. Tea swaps are bespoke
arrangements, wherein market participants exchange their exposure to variable market prices against
fixed prices that they either receive from a broker (if they are a tea seller), or pay to the broker (if they
are a tea buyer). But as long as the broker cannot take on considerable risk – as is the case now – the
growth of the tea swap market will be seriously constrained.
In conclusion, one may agree that “tea producers and exporters are plagued by repeated cycles of price
and revenue fluctuations, endangering the sustainability of a significant part of the industry”. If this is
the case, it should be recognized that while futures contracts are not a panacea, they are effective,
proven, low-cost solutions for dealing with price and revenue fluctuations. It thus makes sense for the
tea industry to act as champion of tea futures contracts, and formulate action plans for each of the
relevant tea market segments.

5
INTRODUCTION
Tea producers and exporters are plagued by repeated cycles of price and revenue fluctuations,
endangering the sustainability of a significant part of the industry. Climate change will only add to the
sector’s insecurity.
FAO’s 71st Session of the Committee on Commodity Problems, October 2016, has underlined the
fundamental role of agricultural commodity prices for sustainable development, especially for
commodity dependent developing countries. Referring to the conclusions of the just-concluded Fourth
Ministerial Meeting on Longterm Commodity Price Trends and Sustainable Agricultural Development,
the committee urged policy makers to give due attention to price movements as they affect adoption of
innovations, rural poverty, food security and structural transformation. 1 The FAO Intergovernmental
Group on Tea (FAO – IGG/Tea) 2, as well as industry bodies and governments, have identified tea futures
as a possible solution.
Efforts to reduce tea price movements through international market intervention have so far not borne
fruit.3 As an alternative, participants in the tea market can be empowered to better manage their
exposure to price movements. Futures contracts in tea, if feasible, would be a useful risk management
(hedging) instrument for producers to cover themselves against price risks. 4 Furthermore, a tea futures
exchange will not only provide a platform for risk management, it will also improve price discovery in the
tea market, which in particular favors the weaker parties in the tea value chain. Beyond their role of
giving farmers more predictable revenues and traders and processors better manageable profit margins,
tea futures have the potential to unlock finance for growers, so that they can better invest in future crop
production.
Through a futures contract (in its simplest form), a market participant agrees to buy/sell a standardized
commodity at a fixed price to be delivered at a specific location at a future date. Delivery normally does
not take place, instead contracts are closed out financially – for a grower, this means that if prices are
lower than anticipated when she entered into a short position, then the lower price for her crop is
compensated by a profit on the futures transaction (in return, the grower loses the potential gain from
higher physical market prices). A futures market will not offer each market participant a hedge for the
exact price risk of his crop, but rather, by providing a standardized industry-wide tea futures contract,
will provide participants the possibility to hedge against general tea market price risk. Exchanges achieve
standardization by specifying parameters such as contract maturity date, trading and delivery lot size,
delivery norms related to packaging, labeling and detailed quality parameters. If the price behavior of a
participant’s specific tea is too remote from that of the standard contract, then the contract is of no
hedging use.

1
FAO, 2017.
2
FAO, 2012a, 2012b.
3
The latest effort was by the Governments of India, Kenya, Indonesia, Malawi, Rwanda and Sri Lanka, who in 2013
created the International Tea Producers Forum, inter alia to stabilize prices. The Forum has, however, been unable
to agree on concrete measures to achieve this goal.
4
For a description of tea futures as they would apply to the tea market, please refer to FAO 2012a, 2012b and
2012c.

6
While there are mixed experiences in agricultural futures markets at the country level, the tea industry
has few precedents upon which to draw for the design of a framework for tea futures trade. E xcept for a
brief and unsuccessful experience in the London Produce Exchange during the last decade of the 19 th
century, tea has not been traded on futures exchanges anywhere in the world. In contrast, futures
contracts for coffee have been actively traded since 1882 and contracts for cocoa since 1925, alongside
many other agricultural commodities such as cotton, rubber, palm oil, wheat and soybeans.
Some in the tea industry believe tea is too complex to trade on a futures exchange. Tea is
heterogeneous both over season and region, and even intra-region depending on the stock from which
harvest was made and the periodicity of the harvest. Thus, there is no obvious “reference variety” that
can act as a standard. They also believe that one cannot readily define a quality for delivery against a
futures contract, or a system of discounts and premiums as exists in the coffee futures market. And if
one were to make a range of tea varieties deliverable against a futures contract, it would be so broad
that it makes the concept of a standardized contract meaningless; for example, in the early 2000s, t he
Tea Board of India proposed a framework for developing commodity based futures contracts in tea
based on more than 50 grades.5 This complexity is a challenge that has to be overcome to create a
viable tea futures contract.
In most countries with organized agricultures futures exchanges, the low participation in futures
markets of farmers, especially small ones, remains a challenge. In general, only large farmers and those
organized in effective cooperatives trade directly on futures exchanges, although if options are offered
as well – these are products akin to price insurance instruments – they tend to be used more widely. 6
Rather, the main benefits of futures exchanges for farmers tend to be indirect: greater price
transparency, which may allow them to make better production and marketing decisions, and negotiate
better sales prices; and buyers offer a broader choice in contracts, allowing farmers greater freedom in
deciding when to fix their prices.
But both these indirect transmission mechanisms can only work well if certain minimum conditions are
in place. For example, even if there is a futures exchange, a lack of adequate warehousing and other
hard and soft infrastructure (handling, transportation, trading, grading and standardization) facilities
may still force farmers to sell during harvest period at low prices.
It is in the context of such mixed expectations that several tea producing countries have envisioned to
create futures markets for tea to address price volatility and its impact on tea farmers’ income. The aim
of this paper is to discuss the pertinence of tea futures market platforms, and assess countries’
readiness and risks as they envision to put in place such demanding structures.
This CRS document aims to support discussions on tea futures and swaps contracts during the 23 rd
Session of the FAO – IGG/Tea. The document briefly highlights the situation of the tea market and
underscores price volatility as one of the major challenges affecting such a highly segmented market
5
FAO, 2012b
6
There are several reasons for this, including the fact that to be cost-effective, delivery lots on futures exchanges
tend to be equivalent to truckloads, too large a volume for small farmers to deliver; and the requirements of
futures exchanges that those buying or selling futures contracts have to deposit “good faith” money (margins)
which may have to be topped up in response to price movements – small farmers rarely have to financial flexibility
to deal with such margin requirements. See for a discussion UNCTAD (2002), Farmers and farmers’ associations in
developing countries and their use of modern financial instruments, UNCTAD/DITC/COM/35, 10 January.

7
before discussing conditions towards successful tea futures and swap contacts. Chapter 1 starts with an
overview of market structures and marketing arrangements in main tea producing countries, as they
affect risk exposure and the ability to control price risks. In this chapter, existing mechanisms through
which the industry mitigates/manages such risks in the absence of a futures market are described, with
special mention to small tea producers. Then, tea price trends and volatility are discussed for different
segments of the tea market, and the implications for tea producers’ incomes and their ability to invest.
Chapter 3 considers in more detail the segmentation of the international tea market, to determine
whether in principle, the global market could be served by one futures contract or not.
Chapter 4 sets out the conditions for the successful creation of tea futures markets; or, as an alternative
instrument, the creation of an over-the-counter or organized system for the exchange of payments
(swaps) to provide protection against price risk. Note that the focus is on the practicalities of
establishing a tea futures markets, not on the principles – for the principles, one may refer to the
abundant literature that has been produced on the functioning and impact of commodity exchanges in
developing countries by in particular the African Development Bank, UNCTAD and the World Bank. The
chapter further discusses to what has happened with earlier efforts to create futures exchanges and
how lack of success so far can be linked to the absence of the necessary pre-conditions in the various
countries. Chapter 5 continues this discussion, and building on the findings of the earlier chapters,
argues how, faced with these challenges, one could still move forward with the establishment of a
futures market. This chapter also discusses “tea swaps”, which have been suggested as an alternatie risk
management instrument. This chapter will describe the concrete actions that would be necessary, in
terms of development of market infrastructure, legal and regulatory framework, and awareness-raising
and capacity development. The sixth chapter discusses, assuming a tea futures market is created, how
tea growers can be enabled to fully benefit from this market. A final section concludes.

8
 1. MARKET STRUCTURES AND MARKETING ARRANGEMENTS IN THE MAIN
PRODUCING COUNTRIES

Production, consumption and trade

Tea is produced in some forty countries with tropical and sub-tropical climates. It has to grow at
relatively high altitude: if temperatures become high the plant starts to wilt, but when grown too high
(above 2,000 meters) there is a risk of frost that kills the plant. The major producers are China, India,
Kenya and Sri Lanka. In 2016, China represented 43% of worldwide tea production while India, the
second largest producer, accounted for 22%. Globally, the tea sector was valued at US$ 14.7 billion in
2016, and 5.46 billion US$ in exports. The sector employs more than 13 million people worldwide – 4
million workers and 9 million smallholders. 7
Fig. 1 – Shares in World Tea Production, 2016 (out of total volume of 5.729 million tons)

Sri Lanka
5%
Rest of the World
22%

Kenya China (Mainland)

8% 43%

India
22%

Source: based on FAO statistics

When considering international tea trade, Kenya is the largest exporter, accounting for a quarter of
global tea exports (in volume terms – in value, Sri Lanka is the largest exporter). It is followed by China,
and then Sri Lanka and India. Unlike China and India, Kenya and Sri Lanka only consume a small amount
of their teas and thus their production is mostly exported. China is the world’s largest tea consumer,
followed by India. Global demand for tea has seen healthy annual growth, and this is expected to
continue. Tea imports are widely distributed across mainly non-tea producing countries (tea producers
import some tea for blending purposes).

7
IISD, 2014.

9
Fig. 2 – Shares in World Tea Exports, 2016 (out of total volume of 1.75 million tons)

Sri Lanka
16%

Rest Of The World

27%

China (Mainland)
19%

Kenya
25% India
13%

Source: Based on FAO statistics

Fig. 3 – Tea Domestic Consumption versus Exports for Major Tea Producers, 2016 (1,000 tons)

Kenya

India

China (Mainland)

Sri Lanka

0 250 500 750 1000 1250 1500 1750 2000 2250

Exports Consumption

Source: Based on FAO statistics

10
Fig. 4 – World Tea Consumption, 2016 (out of total volume of 5.53 million tons)

Sri Lanka And Kenya

1%

China (Mainland)
Rest of the World 39%
41%

India
19%

Source: Based on FAO statistics

Fig. 5 – World Tea Imports, 2016 (out of total volume of 1.84 million tons)

Pakistan
9%
UK
7%
Russian Federation
9%

United States
Others 7% China
Sri Lanka, Kenya,
64% and India
3%

Source: Based on FAO statistics

11
Implications for a tea futures market. The overall size of the tea market is large enough to support a
viable futures contract (the ballpark figure often used in the futures industry to decide whether a
futures market is worth considering is US$ 2 billion), but the futures contract will need to address a
significant part of the market. The wide distribution of imports, with no country leading the market,
indicates that a futures contract, if feasible, has to be structured around the tea producing countries.
Prima facie, there could be a case for domestically-oriented contracts in China or India, and for globally-
oriented contracts in Kenya or Sri Lanka. While it is likely that an effort to create a purely international-
trade-oriented futures contract in China or India will fail (the weight of local demand is such that it
would overwhelm any possible export-oriented contract), a domestically-oriented contract in China or
India can become relevant for global markets if these countries’ tea markets are sufficiently well-
integrated into global tea marketing and pricing.

Tea Processing and Varieties

All varieties of tea are produced from the same plant. What differentiates them is the way that the
leaves are processed. Tea processing involves basic steps which are plucking, withering, rolling, oxidizing
and firing. The varieties of tea are determined by the oxidizing process. One estimate is that there are
more than 3,000 kinds of teas, grouped in five main varieties: White, Green, Oolong, Black and Pu Erh. 8

Table 2: Tea Procedures Producing Countries

VarietiesXVarietie
s of Tea
White Made by plucking and being left to China, Sri Lanka and Kenya
dry.
Green Made by plucking, withering and China, India, Sri Lanka and Kenya
rolling. Green tea is not fermented:
oxidisation is prevented by heating
the leaves. This can be done by
steaming or pan-frying the tea.
Oolong Partially fermented tea, made by China
carrying out the five basic steps with
addition of repeating rolling and
oxidisation process.
Black Made by carrying out the five basic China, India, Sri Lanka and Kenya
steps and it involves complete
oxidisation.
Pu Erh Similar steps to green tea but before China
drying the leaf, it is left to age for
months or years.

8
Tea Class, n.d.

12
Black tea can be further classified into orthodox and CTC (Crushing, Tearing and Curling) tea. CTC
reduces processing costs and produces a uniform output, while the orthodox method maintains the
authenticity of the tea leaves. Black tea produced by the orthodox method carries a price premium. 9
Nevertheless, CTC now accounts for the larger share of black tea production.
Black tea and green tea account for more than 90 per cent of production. Both black tea and green tea
production and exports have been growing, but green tea at a much faster rate. It is expected that by
2023, black tea production will reach 4.17 million tons, and green tea production 2.97 million tons. In
terms of international trade, black tea dominates – out of the 1.8 million tons of tea traded in 2016, only
372 thousand tons was green tea. 10 China is the world leader in green tea production and trade, and has
also seen (and will continue to see) the fastest growth of both tea production and consumption (driven
by rising income levels). Of its domestic production, about two thirds is green tea, and in 2016 (with
exports of 272 thousand tons) it had a 73 per cent market share in global green tea trade. India, Kenya
and Sri Lanka primarily produce and export black tea. In India, CTC accounts for 90 percent of
production, whereas Sri Lanka specializes in orthodox black tea. Kenya is the largest supplier to the
international CTC black tea market. By building its “Ceylon tea” brand, Sri Lanka has been able to
capture much of the value added that normally goes to the companies that blend various tea origins into
a consumer brand. India’s production is mostly of the CTC black tea variety.
CTC black tea is generally used for blending (orthodox teas are also added in smaller proportion to
blends to improve their taste). This makes CTC teas interchangeable, and thus it is likely that even if
absolute price levels will be different from one origin to the other and among grades (CTC teas have own
grading system), overall price movements are correlated as blenders will change the composition of
their blends in response to price changes. Orthodox teas are more diverse, although there are accepted
grading methodologies (with different names used across countries); in the main grading system, there
are a dozen or so main grades for whole leaves, and another dozen or so for broken leaves. 11
Even if black tea and green tea are different products, when production and/or demand can readily shift
from one to the other, price behavior in the two markets can become correlated. On the consumer side,
a preference for black or green tea is mostly driven by health considerations, rather than by price. So
demand factors are unlikely to drive black tea and green tea markets together.
However, producers can and do, to some extent, shift their production from one variety to another. For
example, one estimate for South India (faced with a shift in demand from black to green tea) was that
smallholders could convert 10-15 per cent of their production from black tea to green tea. 12 Smallholder
producers here and in other parts of India indeed did shift to green tea, in response to a faster growth of
demand. However, such a change requires a change in cultivation practices and investments in
machinery. These changes are not easy or rapid, let alone automatic (just in response to price signals).
When demand for black tea in China increased, the country’s green tea producers did not change their
production, but rather, the country started importing black tea.

9
Tea Class, 2017
10
Tea market review for 2016 and outlook for 2017, Daily Mirror (Sri Lanka), 16 January 2017.
11
See https://en.wikipedia.org/wiki/Tea_leaf_grading, and also Victoria Bisogno, BOP, OP, OF, GFOPS… Why are
tea grades important? Worldteanews, 31 March 2014
12
Subramani, M.R. (2012), South to brew green tea to meet rising demand, Business Line, April 27

13
Like for other commodities, there is a growing consumer demand for tea that is certified as meeting
certain sustainability criteria (“standard-compliant”). The main such standards are Fairtrade
International, Organic, Rainforest Alliance/Sustainable Agriculture Network and UTZ. In 2012, 12% of
global production was standard-compliant – as much as 40% of Kenya’s tea, and 18% of India’s tea. By
2015 18.9% of global tea was produced on areas certified by one of the four major standards. Tea that is
standard-compliant receives a premium, of 1 to 20 per cent. 13
Implications for a tea futures market. Despite both green tea and black tea being produced from the
same plant, in practice the two markets appear segmented, with neither producers nor consumers able
or willing to shift rapidly between the two in reaction to changing prices. This implies that it is unlikely
that a black tea futures contract would be representative for green tea prices, or vice versa.
Furthermore, there are considerable differences between countries in the varieties of tea produced and
traded. Thus, a tea futures contract in China would normally be for green tea, and this contract would
have little relevance for the other producing countries. In India, Kenya and Sri Lanka futures would be in
black tea (possibly relevant for about one third of the Chinese market). A Sri Lankan futures market
would be for orthodox tea, relevant for the other countries only if there is a strong price correlation
between orthodox and CTC teas (it is not a problem if absolute price levels are higher – what matters is
just the price correlation).
However, it remains to be seen whether within separate groups (e.g., orthodox teas), the various grades
show some degree of price correlation. It should, however, be noted that if there is a positive price
correlation but it is low, this may be due to the absence of a transparent reference price; once there is
such a reference price, the various segments of the tea market are likely to align themselves to this
reference price, as long as the underlying physical markets for the tea grades are linked (which is the
case for export-oriented markets: the large buyers are active throughout the world).
With respect to standard-compliant tea, as long as prices are expressed in terms of a premium over a
market reference price as they usually are 14, such tea can be hedged on a futures exchange. Thus, the
efforts of the tea sector to move to certified teas does not negatively impact the feasibility of a futures
contract.

Tea Producers
Tea is produced by smallholders as well as plantation estates. Over the years, smallholders have become
more important – “it’s estimated that about 70% of global tea production is from 8 million smallholders
in Africa and Asia”.15 In India, the share of smallholders has increased from 27% in 2010 to 47% in 2017
(in terms of green tea leaves; their share in the production of made tea, which is the product sold at
auctions, is only 34.7%). In Sri Lanka and Kenya, 70% of production is by smallholders 16, and in China,
80%.
13
IISD, 2014.
14
Fairtrade, for example, annually sets a fixed premium over “the normal commercial price”. However, Fairtrade
also guarantees a minimum price, different from country to country. (IISD, 2014) If this minimum price is close to
current market levels, this would discourage producers from managing their price risk.
15
FAO, n.d. Others put the number of smallholders at 9 million.
16
Ethical Tea Partnership, http://www.ethicalteapartnership.org/category/key-areas/smallholder-tea-farmers/

14
The shift from estates to smallholders is a result of tea production economics. 17 In particular, wages
account for a major part of the production costs, and these have been increasing over time for
plantation owners due to stronger regulatory requirements. On the other hand, cultivating tea provides
smallholders with stable earnings, at a relatively low investment. But a trend of labour migration to
cities may reduce the supply of cheap labour to smallholder farms and thus constrain future production.
Estates tend to be vertically integrated. They own and operate factories, are able to sell on the auctions,
and often have links with international buyers that permit them to export directly.
Small tea farmers, on the other hand, can be in a weak position. They have a limited choice of buyers,
among other things because there are not many factories within acceptable distance. They are poorly
informed of prices and market conditions. In Kenya, the government has remedied this through the
Kenya Tea Development Agency (KTDA). Through KTDA, Kenyan farmers are part-owners of the tea
processing plants. KTDA processes and sells most of smallholder’s tea.
After harvesting, tea has to be processed quickly (within hours) and since smallholders do not have their
own processing machineries, they sell their green leaf directly or through collectors to Bought Leaf
Factories (BLF) or to estate factories. The price of leaf smallholders receive tends to be based on the
auction prices obtained by these factories. However, in practice, these factories may not properly
disclose these prices. Furthermore, they may sell high quality tea privately and sell only poor quality tea
through auction.
Implications for a tea futures market. A futures market provides price transparency, price information
and risk management tools, functions which in general are the most relevant for smallholders who have
a weaker, less informed position in the market place and if they depend on one crop for most of their
cash revenue, are more sensitive to risks. Thus, the shift from vertically integrated estates to
smallholder farmers should make a tea futures market more relevant. Estates normally need futures
contracts less, as they often are vertically integrated with packing and blending companies, and because
they have easier access to other ways to manage price risk (such as forward contracts, or access to bank
loans to overcome temporary cashflow problems). But in past years, the margins of estates have been
under pressure by regulations and social obligations. 18 With narrower profit margins, risk management
becomes more important.

Tea auctions
In 2011, more than 2 million tons of tea was sold through auction centers 19 – 40 per cent of world
production, and more tea than was traded internationally. There are a number of such centers, all in
producing countries (the London tea auction, established in 1679, was closed in 1998). Most auctions
operate 1-2 days a week, with sometimes several auctions taking place concurrently (the Colombo Tea
Auction, for example, operates concurrently in three auction rooms). Table 3 gives an overview.

17
Herath & Weersink, 2006
18
According to India’s Tea Board, the cost of production on estates ranges from US$0.38-0.40 (INRs25-30) per kilo,
whereas smallholders can sell tea profitably at US$ 0.23 (INRs15) per kilo. (Dan Bolton, India’s Tea Board Explores
Expanding Tea Auction, Worldteanews, 5 July 2017)
19
IISD 2014

15
Table 3: overview of main tea auction centers

Popular Main Auction Place Website for Auction Centre

Countries Teas
Sri Lanka Ceylon tea Colombo. Trades almost all the http://teasrilanka.org/tea-auction
country’s production.
Kenya Kenya Black Mombasa. Only engaged in export https://www.eatta.com/index.php/
tea trade; accounts for 75% of exports. the-mombasa-tea-auction
India Assam, Electronic platform, used by all seven https://www.teaauction.gov.in/Ho
Darjeeling auction centres -  (Kolkata, Siliguri, me.aspx
and Nilgiri Guwahati, Cochin, Coimbatore,
tea Coonoor and Jalpaiguri). 20 500,000
tons out of the 1.3 million tons
produced in India is traded through
the auctions.21
Bangladesh Chittagong (mostly for domestic -
market)
Indonesia Black tea Jakarta. Trades 22% of the country’s -
(orthodox production.
and CTC)
Malawi Limbe -
Source: Authors’ data.

In terms of volumes traded, the Mombasa Tea Auction is the largest (the Colombo auction trades a
higher number of lots). Mombasa provides a platform for warehousing, handling and overseas shipping.
It acts as an auction platform not only for local producers, but also for other African nations such as
Malawi, Zimbabwe and Rwanda; and occasionally, even from remote origins like Indonesia and Papua
New Guinea. Buyers from more than 50 countries are active on the auction (EATTA, n.d.). Malawi’s
Limbe Tea Auction is also used to sell tea from other countries – Mozambique, Zambia, Zimbabwe.
Factories send their tea in chests or packets to the auction, where they have a broker appointed to act
as their representative. There are only a small number of brokers on each auction. The tea is catalogued
by the tea auction organizer, and stored with a warehouse keeper. Prior to the auction, the tea will be
tasted by interested buyers – tea quality can vary considerably even from the same factory or region. 22
The organizers are responsible for scheduling the sales, handling the auctioning process, monitoring
auction activities and releasing sales reports.

20
In terms of relative importance, traded volumes (in million of kgs) in 2013 were as follows; “Tea Serve” refers to
tea that was offered Guwahati
Kolkata
163 directly
127 on the electronic
100 platform,Coonoor
Siliguri not60through oneCochin
of53
the auctions.
Coimbatore
18 Tea16Serve

21
Officially, since October 2015 growers are required to sell half of their tea at auction.
22
Groosman, 2011. There is also a small auction center for green tea in Amritsar.

16
The auction organizers will determine the initial opening auction price based on assessment of potential
demand and the bidding behavior on the day, and also taking into account the previous auction’s prices
for similar tea. There are only few bidders; most represent major tea companies, and there is potential
for collusion. Furthermore, tea companies will generally bid on the tea produced by their affiliated
factories (contrary to an exchange, trade is not anonymous); this keeps other buyers away. Tea is sold to
the highest bidder.23 Warehouse keepers will keep the tea on their warehouses and allow buyers to
collect their teas, through permission receipt issued by brokers. The seller receives the bid price, minus
broker’s commissions, warehouse fees and taxes. 24
The main problem of the auctions is that their processes are very slow. In Mombasa, for instance, trades
take 45 to 60 days to conclude; even once the auction has become electronic (as is planned), it will still
take a month. In India, waiting time in the Northern auctions increases strongly for teas that enter the
auction towards the end of the harvest season.
Apart from the auctions, there is also organized tea trade in the Dubai Multi Commodities Centre
(DMCC), representing 1-2 per cent of world trade. 25 Tea from 13 countries were, that year, traded
through Dubai. The DMCC Tea Centre offers warehousing, blending and packaging facilities. DMCC owns
a majority share in the country’s futures exchange, which has in the past considered the introduction of
a tea futures contract, but without coming to a positive conclusion as to its feasibility. 26
Implications for a tea futures market. At times, tea auction centers see a futures exchange as a
competition to their business. However, the two address different purposes. The auction provides a
mechanism that permits the best possible price to be achieved on a given day for a given tea origin. The
futures market provides information about the general trend of prices, and makes it possible for
producers, factories, traders and buyers to lock in future prices. The two can complement each other,
even explicitly – for example, instead of the auction being for absolute prices, it could be for premiums
and discounts, with the bidder offering the best premium over the day’s futures price being the winner.
Nevertheless, the nature of the auction, with sellers and buyers focusing on the specifics of each
parcel/chest, can easily give rise to the idea that each tea is unique and therefore, a market that relies
on standardization cannot work… this perception is erroneous, and if a futures contract appears feasible
(to be further explored in the remainder of this paper), then building strong links with the relevant
existing auction centers and their main actors, and creating awareness about the complementarity of
the two systems is imperative for its success.

23
The tea that remains unsold will be made available at the next auction (Hossain et al, 2016).
24
Hazarika, n.d.
25
One article (Dubai records spiral growth in tea trade driven by China stimulus, Cayman iNews, 10 April 2016)
quotes the DMCC Tea Center Director reporting a 2015 volume of 41,000 tons, which is more than 2 per cent of
world tea trade. But on its website, DMCC reports that US$ 48 million worth of tea passes through the country,
slightly more than 1 per cent of world trade.
26
For example, Dubai wants to launch world’s first tea futures contract, Daily News (Sri Lanka), 21 February 2008;
and again a few years later, Dubai commodity exchange in talks with producers to launch tea futures, Economic
Times, 16 September 2015.

17
The Tea Value Chain
Most of the plucking of tea is done by women (men tend to do the pruning, planting, and application of
fertilizers or pesticides). Tea plucking takes place throughout the year, although there is some
seaonality, both to avoid the extra labour costs that come with harvesting when the soil is waterlogged
(during the rainy season), and to optimize tea quality. Once the leaves are picked, they have to be
processed. Unless if they are part of a cooperative with its own factory (which is rarely the case outside
of Kenya), they sell their leaves either to an independent factory or an estate. In Kenya, the Kenya Tea
Development Agency (KTDA) is owned by the country’s around 560,000 smallholders. KTDA operates the
tea factories, and when farmers deliver they are paid an advance price, fixed on an annual basis. The
factories then sell the tea, mostly through the Mombasa auction. Once a year, farmers are paid a bonus
depending on the prices realized for the tea sale.
After the tea has been processed, it can be traded, either domestically or for exports. Brokers play a
large role in the process, linking buyers and processors. The major part of the trade is through auctions,
and the remainder is within a vertically integrated company, or through private deals facilitated by
brokers. Table 4 describes the largest companies active in the tea chain.
Table 4: main companies in the different parts of the value chain

Producing & processing Trade Blending & packaging

Unilever X X (buying from 750,000 X (11% of global
smallholders) market)
Tata Global Beverages X (annual production of X X (4% of global market)
70,000 tons)*
Mcleod Russel X (world’s largest tea
producer, producing
over 100,000 tons of
black tea a year)
James Finlay X (producing around X (trades around 70,000
60,000 tons a year) tons a year)
John Keells X (major estate
company in Sri Lanka)
Van Rees X (trades around 80,000 X
tons a year)
Associated British Foods X (3% of global market)
Tenfu Corporation X X (internationally X (China’s largest tea
active) seller)
Zhejiang Tea Group X X (with 40,000 tons, X
world’s largest green

18
 tea exporter)
Hunan Tea General X X (buying from 400,000 X (selling 50,000 tons a
Corporation farmers) year)
Source: Based on Groosman, 2011; with information on specific companies added by authors, based on
various reports. *In 2007, Tata started a process to withdraw from plantations. In December, Tata sold
its stake in a major Sri Lanka tea plantation company; and as of January 2018, Tata was considering the
sale of its major plantation company, Amalgated Plantations, which produces 26,000 tons a year (and
processes a further 17,000 tons bought from smallholders).

Most of the value addition in the tea sector is in the downstream parts of the value chain, as illustrated
in Figure 6 (representative of tea sales in the UK in the late 2010s). Note that this may give a relatively
negative impression of tea pickers’/growers’ share of the final tea value – in other cases, farmers
reportedly receive a higher share of the final consumer price.
Figure 6: distribution of value in the tea value chain

Tea Trader/
Auction/
picker/growe → Buying → Factory → → Blender → Retailer
broker
r agent
1-3% 6% 7% <1%* 33% 53%
Source: Based on Anna Morser, 2010. Auction fees are charged on the basis of the prices realized at the
auction. In Kenya, the seller pays 1% and the buyer 0.5%. In India, the commission is around 2%.
With respect to consumer demand, there is a trend of “premiumization” in the tea market – both in
developed countries and in the major tea producing developing countries. 27 Consumers are willing to
pay more for premium teas. To some extent, this will lead to markets for certain specialty teas becoming
independent from the tea market as a whole (in China, for example, there are teas for which consumers
are willing to pay over hundred times the price of ordinary tea). But probably (this deserves further
exploration), most specialty teas will be priced at a premium to “wholesale” teas, and thus its prices will
remain linked to the general tea market.
Implications for a tea futures market. For a futures market to be viable, there should be a competitive
market, with no company able to unilaterally dictate prices. In this respect, there are no issues in the
production and trading segments, nor is there a major problem with intra-company trade. However, the
market share of a few large multinationals in blending can be problematic. The larger blender’s share of
11-12 per cent of the world market in fact means that it accounts for between a quarter and a third of
international tea trade. This means that if it does not wish an international tea futures market to
succeed, it can arrange for its failure. Thus, if such a market is deemed feasible, it is important to engage
the large blenders at an early stage to ensure that at the very least, they do not oppose such a market.
The distribution of value in the chain points suggests a possible obstacle. If the standard tea for a
futures contract is defined as ex-factory (“made tea”), the price of this tea is, on the basis of the above
numbers, at best (if the smallholder sells directly to the factory) around double the farmer’s price, and

27
FAO, 2018.

19
less than a third of the price that the blender receives from the retailer. This increases the likelihood
that price movements at each level are not well-correlated, making the futures contract less useful as a
hedging tool for farmers and for blenders (while the latter could hedge their purchase prices, they may
not be able to hedge their sales prices to retailers).

2. TEA PRICES AND VOLATILITY

The chapter starts with an overview of tea price developments, followed by a discussion on tea price
seasonality and volatility.

Tea prices
Figure 7 shows tea price movements (based on the average auction prices of Kolkata, Cochin, Colombo
and Mombasa) for 2000 to 2017.
Figure 7 – FAO composite tea price, 2000-2017

3.50

3.00

2.50
Price (USD/KG)

2.00

1.50

1.00
00 00 01 02 02 03 04 04 05 06 06 07 08 08 09 10 10 11 12 12 13 14 14 15 16 16 17
a n- ep- ay- an- ep- ay- an- ep- ay- an- ep- ay- an- ep- ay- an- ep- ay- an- ep- ay- an- ep- ay- an- ep- ay-
J S M J S M J S M J S M J S M J S M J S M J S M J S M

Date

Source: based on FAO data

Figure 7 gives a good idea of the overall development of tea prices, but for the purposes of this
study it is useful to consider how prices in the main auction centers have performed. Figure 8
shows the data (with Coonoor replacing Cochin as in the last year for which data were available, it
was the bigger auction), using weekly auction data for the Indian auctions and monthly ones for the
two other ones. Prices very much suggest that the four markets are distinct.

20
Figure 8: Kolkata, Coonoor, Mombasa and Colombo tea auction prices, 2014-2017

Source: based on auction center data:

- India: http://www.teaboard.gov.in
- Sri Lanka: http://teasrilanka.org/tea-prices
- Kenya: https://ycharts.com/indicators/kenya_tea_auction_price

21
Tea price seasonality
As is shown in table 5, on the aggregate, tea prices show modest, but clear seasonality. Prices in May
and June tend to be below the average for the whole calendar year, and prices from July to October
tend to be higher. In both cases, this is true for at least two out of three years.
Table 5: Seasonality of FAO composite tea prices, 2000-2017
Month Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec
Average difference
(2000-2017) from each -0,7% -1,9% -0,4% -1,4% -2,1% -1,8% 2,2% 2,1% 3,2% 1,9% -0,3% -0,9%
year’s average price

Number of years that

month’s price was 9 9 8 10 14 13 7 6 6 6 9 9
below year’s average
Number of years that
month’s price was 9 9 10 8 4 5 11 12 12 12 9 9
above year’s average

Source: calculated from FAO data

Figure 8 clearly shows strong seasonality of the Kolkata auction prices (no such seasonality can be seen
for the other origins). Prices hit a peak in August-September, and hit bottom in May. Remarkably, peak
prices are as much as double of bottom prices. Sarkar, 2012, notes that ”there is a clear seasonality in
prices of tea within a year. The arrival of the first flush tea in April normally causes prices to rise, which
then decline during July with the arrival of the monsoon, and again rise in October, with autumn flush
fetching better prices, and fall in January. This pattern of seasonality is seen in most of the auction
centers.
Implications for a tea futures market. Seasonality is good for futures trade. In the case of tea, factories
that sell (at a price reflecting the composite price levels) in May or June – because that is when their tea
is available for sale – can “unprice” their sales by buying nearby futures contracts and selling contracts
for the July-October months. In that way, without affecting in any way their normal sales practices, they
can receive the above-average prices for July-October rather than the below-average ones for April and
May. Note, however, that seasonality of auction prices may just be a compensation for the seasonality
in moisture level of the tea leaves – with tea leaves having a higher level of moisture when picked during
the rainy season, they are likely to show lower prices than tea leaves picked during the dry season (why
would buyers pay for the water caught in the leaves?). If this is the case, then this seasonality may well
disappear for a standardized tea futures contract.

Tea price volatility

Figure 9 gives an impression of tea price volatility as compared to cocoa and coffee. Until 2007 tea
prices were visibly less volatile, but over the past ten years price behavior has come to resemble that of
the two other beverages – even though it remains somewhat less volatile.

22
Figure 9: behavior of tea prices compared to cocoa and coffee prices, 2000-2017

Source: Tea prices from FAO; coffee and cocoa prices from tradingeconomics.com
But while tea price volatility may have increased, has it become a significant factor for tea sector
operators? A good measure of tea price volatility in terms of its real-world implications is what it means
for the predictability of future realizations. How much do prices normally change say, from month to
month? Tables 6 and 7 shows the results for the Kolkata and Mombasa auction markets (Annex A has
the numbers for the other Indian auction centers as well as Sri Lanka).

Table 6. Frequency Distribution of Price Table 7. Frequency Distribution of Price

Changes of Kolkata Average Weekly Auction Changes of Mombasa Average Monthly
Prices, 2008-2017 Auction Prices, 2013-2017

% Price
Frequency Frequency Frequency Frequency
% Price changes changes
Distribution Distribution Distribution Distribution
compared to compared
(Increasing (Decreasing (Increasing (Decreasing
previous month to previous
price) price) price) price)
month
0-5% 32% 48%
5-10% 5% 8% 0-2% 8% 10%
10-15% 1% 1% 2-4% 18% 8%
15-20% 0% 0% 4-6% 6% 8%
20-25% 1% 1% 6-8% 6% 10%
25-30% 1% 0% 8-10% 10% 6%
30-50% 0% 0% >10% 6% 2%
>50% 1% 0%

23
The tables mean, for example, that based on past data if a grower whose tea is sold in Kolkata observes
a good price on the auction market and decides to pick more in response, there is a 10 per cent chance
that by the time the tea is priced the next month, its price has fallen by more than 5 per cent; and a 2
per cent chance that it has fallen more than 10 per cent. For a Kenyan grower (based on monthly
averages28), there is an 18 per cent chance that by the time the tea is priced the next month, its price
has fallen by more than 6 per cent; and a 2 per cent chance that it has fallen more than 10 per cent. For
a Sri Lankan grower, there is, based on past data, a zero chance that prices will fall by more than 6 per
cent. Even the higher numbers for Kenya do not indicate a market which is so volatile that companies
have to hedge to keep their risks at acceptable levels.
While proper price series for China are not available, average prices change little from year to year (see
Table 8).
Table 8: Tea producer prices in China, 2010-2016
Year Production Agriculture Output Value of Producer Price Indices of Tea
Tea (Billion Chinese Yuan) (Preceding year=100)
（Million kg)

2010 1,475 67.6 120.38

2011 1,623 74.4 113.34
2012 1,790 91.57 108.78
2013 1,924 105.9 100.47
2014 2,096 123.71 105.49
2015 2,249 130.1 98.66
2016 2,405 136 99.37
Source: data provided by FAO

Implications for a tea futures market. Low volatility discourages the use of futures markets by hedgers,
and also makes a market less attractive for speculators (who are essential for the success of a futures
exchange). Available data in this respect paint a mixed picture. On the one hand, tea prices are volatile
– a bit less than coffee or cocoa prices, but not so much that the market would be unattractive for
speculators. But on the other hand, tea sector participants are exposed to price risk for a relatively short
time frame, given that tea is produced throughout the year and tends to move in a straightforward way
from producer to blender; in contrast, cocoa and coffee harvests are seasonal, and there are large stocks
kept for many months at several levels of the value chain. Within this short time frame, prices do not
appear to move very much; unlike in the cocoa and coffee sector, the various participants in the value
chain do not seem to be exposed to the risk of catastrophic price changes. Thus, they will not be
automatically called to use a new tea futures market – significant efforts would be needed to elicit the
benefits of exchange use and build awareness of these benefits.

28
It is likely that these numbers underestimate volatility: monthly averages smooth week-to-week price
fluctuations.

24
 3. HOW WELL ARE THE PRICES FOR DIFFERENT VARIETIES, GRADES AND
MARKETS CONNECTED?
The core question in this chapter will be “is there one tea market, or are there several tea markets”.
How well are the various origins, varieties, prices for tea that are certified under fair-trade premiums
etc. correlated?
First, market shares for the different varieties etc. will be described. Then, after an analysis of price
correlations, the results will be discussed, in the light of the actual physical aspects of the tea market.
Note that even if the numbers show several separate markets, this does not automatically imply that a
futures market can only serve one part. It is possible (the physical tea market will give the indications for
this, if any…) that price behavior differs from market to market because there is no central price
reference point, and the creation of a futures exchange can change this (this typically has happened
when futures in a food commodity are introduced – before, prices in different cities were poorly
correlated, but once there is a futures contract all markets start behaving the same). Also, there may
be human-made factors that cause markets to be less aligned (e.g., government interventions) that can
be remedied.

Market correlations
Data on tea prices are poor. Notably, proper price series for China are missing. For other countries, there
are price series for the auction centers (at least on a weekly basis); however, these do not differentiate
between grades and varieties of tea.
For CTC teas in India, the price correlation between the country’s auction centers was analyzed. As can
be seen from Table 9, correlation coefficients range widely. Some pairs have relatively high correlation
with a coefficient of 0.8, which are useful for the development of futures contract. With a price
correlation of 0.8 or above, a contract can be efficiently used for hedging. With a price correlation
between 0.7 and 0.8, a contract can be used from time to time to improve price realization, but not to
support a systematic hedge strategy. With a price correlation of less than 0.7, companies would be
increasing their risk exposure by moving their position into the futures market (i.e., it would be
speculation).
Auction centers with high correlation coefficient are Kolkata-Guwahati, Guwahati-Siliguri, Cochin-
Coonoor, and Coonoor-Coimbatore. All other pairs show a low coefficient.

25
Table 9. Coefficients of Correlation of Tea Prices (CTC leaf & all dust) of Auction Centers in India,
2008-2017
  Kolkata Guwahati Siliguri Jalpaguri Cochin Coonoor Coimbatore
Kolkata 1 0,79 0,69 0,40 0,31 0,19 0,27
Guwahati   1 0,72 0,40 0,30 0,18 0,26
Siliguri     1 0,34 0,30 0,20 0,26
Jalpaguri       1 0,18 0,14 0,15
Cochin         1 0,69 0,84
Coonoor           1 0,85
Coimbatore             1

This suggests that for CTCs in India, there are two segmented markets, one in the North (Kolkata-
Guwahati-Siliguri), and one in the South (Cochin-Coonoor-Coimbatore). This is a more negative finding
than a 2012 study of the black tea market 29, which used data from seven auctions, and found that the
markets were segmented between orthodox teas (Sri Lanka’s main export) and CTC teas. It found that
for CTC teas there was effectively one market across countries.
Extending the analysis to other countries, only monthly data are available, albeit for a longer period. The
results are as follows (and annex B shows the scatter charts for the underlying price pairs):
Table 10. Coefficients of Correlation of Tea Prices in the Kolkata, Cochin, Colombo and Mombasa
auctions, 2000-2017
Cochin Colombo Mombasa
Kolkata
2000-2017 0.74 0.72 0.71
2000-2009 0.62 0.57 0.52
2009-2017 0.05 -0.01 0.13

2000-2017 (trend-corrected) 0.52 0.34 0.53

Cochin
2000-2017 0.89 0.79
2000-2009 0.82 0.63
2009-2017 0.39 0.18

2000-2017 (trend-corrected) 0.62 0.58

Colombo
2000-2017 0.78
2000-2009 0.58
2009-2017 0.14

2000-2017 (trend-corrected) 0.43

Source: based on FAO auction data. The trend-corrected numbers use each
auction price series corrected for its linear trend from 2000 to 2017.

29
Tanui, 2012

26
In the long run, prices moved together, but this was partially because they all followed an upward trend.
If this trend is removed, correlations fall. Further, comparing the 2000s and the 2010s, markets seem to
have become much less correlated. This is a somewhat surprising finding as earlier as many of the
international buyers buy CTC teas internationally, and a significant part of this tea is used for blending
and thus to some extent, interchangeable – one would expect higher correlations, in particular for the
markets where CTC teas predominate.

Implications for a tea futures market. Price data confirm that what may look like a global tea market in
fact is composed of a number of uncorrelated markets: in all likelihood, China for green tea; Colombo
for orthodox black tea; and separate CTC black tea markets in North India, South India and Kenya. With
respect to CTC black tea, it is possible that the markets are segmented because there is no central
marketplace, and once this marketplace is created, the various origins will align to its prices (after all,
CTC black tea largely serves the same final market, for blended teas). But until such alignment occurs, a
futures exchange organized around any of these regions will have difficulty attracting hedging volume
from other regions.

Quality correlations
FAO, 2012a, analyzes correlations in India between various grades of teas. It finds, inter alia, that
“higher priced teas have higher variance in weekly averages than the common teas” (suggesting that
teas that are used for blending show more stable prices, which indicates in turn that they are seen as
interchangeable with other similar teas).

Prices for different teas vary widely. Figure 10 gives an impression for Northern India.

Figure 10: Average auction prices (2010) for various tea varieties (INR/kg)

400
350
300
250
200
150
100
50
0
C . ar .) rs i a .) a
CT th ch th oa ra ur th te
or a or o Te ip (o
r n
am m C g ( D Tr
r ee
ss sa lin kim G
A As jr ee Sik
Da

Source: FAO, 2012a

27
The monthly data in FAO, 2012a, suggest that while some varieties (in particular, Darjeeling tea) show
price behavior that is very different from that of other teas, the standard CTC and “leaf and dust” teas
that are traded in the Northern auctions show fairly similar price behavior.

Implications for a tea futures market. Tea that shows price behavior that is much different from that
of other teas should not be included in the basket that forms the basis for a futures contract (whether
for physical settlement or the calculation of an index value). Analysis of price data for specific teas will
make it possible to narrow down the varieties/grades that can be included.

Overall assessment of the scope for tea futures

If one were to rank the prospects for creating a viable tea futures markets on a scale of 1 to 10, with 1
being completely impossible and 10 being very easy, tea futures would score 5 or 6. Close to a failing
grade, but with hard work it may be possible to convert it into a pass.

Of the various segments of the tea market, China’s green tea segment is the largest, worth around US$ 4
billion. That is normally more than enough to support a viable futures contract. However, if the annual
price data of Table 8, showing low year-to-year price changes, are representative for the normal state of
the Chinese market, then there would be no demand for futures: with low price variability there is no
need for hedging, and no interest of speculators to become involved. More detailed price data for 2017
suggest that within the year, there is volatility, however, with in 2017 the lowest green and black ta
indices reported to be about a third lower than the highest index levels for that year.

However, the Chinese market appears so fragmented that it will be difficult to organize a futures
market. As is further discussed in the next chapter, for a futures market to work well it has to offer a
final settlement mechanism that ensures that, at expiry, the futures price equates the physical market
price. This settlement mechanism requires a reasonably organized physical market, either in the sense
of central nodes in tea value chains through which large volumes of tea pass (such as a port, or a major
trading site), or as an electronic market to which a large percentage of physical traders are connected
and physically trade. Neither seems to be present in China. There are large wholesale markets, but
these seem to operate as places where physical traders congregate, but do not necessarily connect (i.e.,
there is no platform in these wholesale markets on which many of these present in the market trade
with each other). So before a futures contract becomes feasible, an effort needs to be made to better
organize the physical market.

Black tea markets are well organized, with auctions not only acting as nodal points for physical trade
flows, but also offering a mechanism through which market participants agree on grades and standards.
A futures exchange can build on this, with physical deliveries organized around the auction
infrastructure, or using auction prices to establish a generic reference price for tea. Unfortunately, black
tea markets are segmented. The only segment that is large enough to readily support a tea futures
contract is that for CTC tea in North India. Prices on the South Indian auctions are poorly correlated with
those in North India, so one would need a separate South Indian CTC tea contract to serve this market.
Orthodox teas (the smaller part of supply, perhaps 20-30%) would be covered by a third contract. These
contracts could be covered on one platform, and perhaps enough of the interest that can be generated
for the US 2 billion North Indian CTC market would spill over into the two other contracts to make them

28
viable (on a stand-alone basis, their underlying markets are much too small). Earlier analysis has also
suggested that such contracts would be viable. 30 Since this earlier analysis, conditions for a futures
contract have actually improved, because of the introduction of an electronic auction platform, the
overall growth of tea production, and a shift from plantations to more vulnerable smallholders.

Indian Orthodox black teas could also be served by a futures contract centered around Sri Lanka’s tea
exports. However, this market segment is somewhat small for supporting a viable market. Furthermore,
the costs of introducing tea futures would be high because there is no futures exchange in Sri Lanka that
could add tea to its product offering. One has to be innovative in reducing the costs and maximizing the
reach of a tea futures contract to succeed in the face of these challenges.

In terms of volume, Kenya exports more than Sri Lanka, but in terms of value its exports are actually less
(US$ 1.3 billion vs US$ 1.5 billion). The country would face the same challenges as Sri Lanka, and
similarly an innovative approach, building on the internet, is necessary to address these challenges. A
commodity exchange is in the process of being created in Kenya, and if this exchange has the right
business model, it could become the carrier for a CTC black tea futures contract. The chances of success
for such a contract will be better if a North Indian contract succeeds and if contract affects pricing in
Mombasa: if the price correlation between Kenyan and North Indian prices improves, arbitrage becomes
possible, which would boost volumes at both exchanges.

Cocoa and coffee futures are traded in the United Kingdom (UK) and the USA – i.e., outside of the main
producing countries. Could tea futures similarly be traded in the UK, USA, or perhaps Dubai as the new
regional trading center? These markets have the trading facilities and access to the speculative
community that can help build liquidity in a new tea futures contract. So it would certainly be possible.
However, to succeed, exchanges in these countries would need to collaborate closely with the industry
in the origin countries to ensure proper final settlement of futures contracts; and they would need to
invest in awareness-raising and marketing in the origin countries. It can therefore be doubted that they
have any competitive advantage over domestic initiatives.

30
FAO, 2012a, refers to a report by A. Damodaran of the Indian Institute of Plantation Management to the Tea
Board of India, in which he “came to the conclusion that such a initiative was indeed feasible since the spot market
had sufficient volume, turnover and liquidity, was adequately regulated, operated under a few centralized
umbrellas, and also had inbuilt quality regulation mechanisms in the shape of brokers being the unspoken
intermediary in every transaction. Damodaran concluded that apart from producers, all three categories of buyers
in the Indian auctions, packeteers, importers agents and bulk tea merchandiser’s agents, would find futures useful.
He further suggested that the contracts be restricted to a few volatile high turnover Orthodox and CTC grades, be
sold through two centres, one in North India and one in South India, be sold over a time cycle of 14 months, and
delivered from “certified” warehouses.”

29
 4. CONDITIONS FOR THE SUCCESSFUL CREATION OF TEA FUTURES AND
SWAP CONTRACTS
In principle, for tea to be a likely candidate for a successful futures contract it needs to meet the
following criteria:
- Sufficiently large supply and demand for the commodity (i.e., it does not help if the market is
highly segmented). In the case of tea, the market is clearly segmented. Green tea has a different
market from black tea, and with only limited flexibility/interest from producers and consumers
to switch rapidly from one to the other in response to price signals, there is little that would
make the two markets move in tandem. Furthermore, in the black tea segment, there is a clear
difference in the physical market for orthodox black teas and CTC black teas. Somewhat
surprisingly, even within the CTC black tea markets, there appear to be three uncorrelated
segments: North India, South India and Kenya.
What was a 15 billion US$ tea market thus includes five major separate markets, of roughly 4
billion US$ (green tea in China), 1.5 billion US$ (Sri Lanka’s orthodox black tea), and for the three
CTC black tea markets –North and South India, and Kenya– respectively 2 billion, 1 billion and
1.3 billion US$, (estimated 2016 values). Only the Chinese green tea market is large enough to
support, in principle, a futures market; the North Indian market is of a borderline size. The three
other market segments are quite small.
- Sufficiently free determination of prices (little likelihood of price manipulation – which is a
matter of industry structure, and thus the dominance of a few large buyers is potentially
problematic). This condition is largely met. In the green tea market, there are a large number of
actors, none of which is dominant. In the black tea market, it is necessary to bring the large
blenders on board to make success more likely.
- Reasonably well-standardized commodity, and accepted grades. It has been argued that tea
does not meet this particular condition 3: there is no “standard”, dominant type of tea; tea
grading remains subjective; and most teas show a deterioration of quality when stored for
longer than 6-8 months), and many industry players feel tea is too complicated a commodity for
futures trading. This remains a problematic area, but more because of industry mindsets than
because of the inherent characteristics of tea. A tea futures contract formulated along the lines
of the Arabica coffee contract, further discussed below, should be able to accommodate a broad
range of grades and varieties.
- Sufficiently large price fluctuations to warrant hedging. Price volatility is high enough to warrant
hedging, but not high enough to make survival difficult for companies that do not hedge. So
companies have to feel motivated to hedge, which points at a need for considerable awareness-
raising and training as a precondition for a successful futures contract.
- Reasonably well-functioning spot market. Except perhaps in China, this does not seem to be an
issue. The auctions function reasonably well, and a futures exchange can build on the auctions’
strength to build a robust futures contract. Information on China is scarce. The country primarily

30
 produces green tea (66% in 2013 31), followed by black tea (10.5%). Oolong tea accounted for
12%. There is one physical location where tea from all over the country is gathered (the Fangcun
wholesale market in Guangzhou, where some 3,000 tea stores are gathered), but it does not
seem there is a mechanism to bring these stores together in one marketing and price discovery
platform. It is likely that in order to successfully establish a green tea futures market in China,
the organization of the physical market also needs to be improved.
- Support from commercial interests for the futures market. Over the past twenty years, private
sector actors in Dubai, India, Indonesia, Kenya and Sri Lanka, have expressed their interest in
seeing a tea futures market established. So at least segments of the market support the idea.
Other segments seem to be skeptical rather than hostile towards the idea of a tea futures
contract.
- Sufficiently large group of speculators. Speculators are necessary for a successful futures
contract because they will provide the liquidity that makes use of the exchange easy and low-
cost for hedgers. On a successful futures market, speculators will account for the bulk of trading
volume. Attracting speculators to the tea market may be problematic. China’s experience with
PuErh tea seems to suggest that speculators can be interested in tea. However, PuErh is not a
typical tea variety; and the PuErh experience also shows that interest in PuErh did not flow over
into speculative investments in other teas. There is speculative interest in coffee and cocoa, but
this is to some extent driven by the seasonal nature of harvests – frosts in Brazil can lead coffee
prices to shoot up, political problems in West Africa can make a cocoa harvest fail. For a crop
that can be produced throughout the year, the drivers for speculative interest are much less
clear. It is, however, well possible that a commodity futures exchange that already has many
speculators among its users (like is the case for the Chinese and Indian exchanges) can convince
part of them to invest in tea, as part of their portfolio diversification.
- Sufficiently well-developed infrastructure (grading, storage, etc.). The auctions all have well-
established warehousing arrangements, and grading arrangements that are understood and
accepted by market participants. To avoid having to replicate this infrastructure, it would be
advisable for an incumbent tea futures market to collaborate with the relevant auctions.
- A supportive legal and regulatory framework. There are some constraints in this respect. India
has a good framework for futures trade, but only domestic users are permitted; it would require
work with the regulator, the Securities Exchange Board of India, to get an exception for tea
futures. China likewise has a good legal and regulatory regime, and it is in the process of
opening up its highly liquid commodity exchanges. Kenya is in the process of establishing a
commodity exchange; but this remains a government-driven initiative, and the legal and
regulatory framework is not yet in place. In Sri Lanka, the decision has been made that a
commodity futures exchange will fall under the aegis of the securities regulator, but a legal and
regulatory framework remains to be elaborated.

31
FAO, 2014.

31
 Box 1 – can tea be interesting for investors? The case of PuErh tea in China.

PuErh is a rare tea that can only be produced in the environment and soil conditions of Yunnan.*
Unlike other teas, this tea actually improves in taste as it is stored. This has attracted investors who
buy the tea for speculative purposes, in the hope of future price gains.

Between 2006 and 2007, driven by a false sense of market scarcity that was created by a number of
traders and tea factories, this speculation became massive. Many small investors poured their life
savings into PuErh tea, which as a result saw a five-fold price increase on China’s main tea wholesale
market, the Fangcun tea market in Guangzhou – it has been estimated that as many as 10 million
people were involved in PuErh investments.* Many tea traders became millionaires.

But the market corrected, partly because supply was larger than many investors seemed to realize
(and growing as farmers were planting new trees), partly because fake PuErh started entering the
market. Once investors realized the price rise was not sustained they started to dispose of their tea,
causing a market crash, with prices falling by 60 per cent in three days. Many investors suffered large
losses.# This was not the first boom-and-bust of PuErh prices – Hong Kong (1950) and Taiwan (1995)
had earlier seen crazes. But investors thought this time would be different…

* Teasenz, 2017
# Jacobs, 2009

Can one learn from the coffee market?

Tea futures have been studied repeatedly in past years, in Dubai, India, Indonesia and Sri Lanka. 32 The
conclusion has always been that it is difficult (although several exchanges have come close to launching
the contract), particularly quoting the lack of uniform qualities. However, this is not a strong argument.
Note the delivery specifications for the Arabica coffee futures contract traded at the Intercontinental
Exchange (ICE): “Mexico, Salvador, Guatemala, Costa Rica, Nicaragua, Kenya, Papua New Guinea,
Panama, Tanzania, Uganda, Honduras, and Peru all at par, Colombia at 400 point premium, Burundi,
Rwanda, Venezuela and India at 100 point discount, Dominican Republic and Ecuador at 400 point
discount, and Brazil at 600 point discount.” To further deal with quality differences within specific
origins, the exchanges tests the coffee: “A Notice of Certification is issued based on testing the grade of
the beans and by cup testing for flavor. The Exchange uses certain coffees to establish the "basis".
Coffees judged better are at a premium; those judged inferior are at a discount.” 33 Coffee is also
deliverable against the futures contracts in many locations: “The New York and Virginia delivery points
are par; the Bremen/Hamburg, Antwerp and Barcelona delivery points are at a discount of 1.25 cents/lb;
and the New Orleans, Miami and Houston delivery points are at a discount of 1.25 cents per pound.”
Coffee futures have been successfully traded since 1882. As ICE, the largest coffee futures market, notes
in its coffee futures brochure: “Any consumer knows coffee is not coffee; there are many grades and

32
Press reports suggest tea futures were also considered in Kenya, but in fact, this was merely an idea being
floated, not based on any study.
33
https://www.theice.com/products/15/Coffee-C-Futures

32
varieties. Still, traders can and do trade one coffee future against another.” 34 To deal with this diversity,
all coffee delivered to the exchange is valued very closely to its actual value on the market. The coffee
itself is not standardized: individual quality differences, even as they differ from lot to lot, are clearly
recognized and valued. However, coffee sector participants have become used to thinking of their
product in relation to a somewhat abstract “standard contract”, rather than existing in isolation. While
it will require a mindset change, a similar mechanism could be feasible for a tea futures contract.
If one prefers to avoid physical delivery, it is possible to have an index contract. Index contracts are not
ideal in the commodity sector: whereas if one has a well-designed physical delivery contract,
discrepancies between the physical and futures market will be automatically dealt with through
arbitrage between the two markets, involving physical delivery into and out of exchange-accredited
warehouses.
No such arbitrage is possible for an index contract, and thus, there is a larger risk of manipulation of
prices. In principle, auction prices are competitive and transparent. However, there have in the past
been investigations into auction pricing that have found instances of manipulation by buyers. So while
the use of auction prices to settle futures contracts may be feasible, an exchange would have to exercise
firm oversight over the price formation process.
In all, while a tea futures contract may be difficult to formulate and cannot be launched without
considerable preparatory efforts, it is probably possible – for green tea in China (but with much work to
be done to improve the transparency and organization of the physical tea trade), and for CTC black tea,
at least for North India. But an exchange will not move ahead with a tea futures without firm and broad
support from the tea sector – given market segmentation, addressable markets are too small, and there
are too many supports that need to be in place (including in terms of awareness-raising and training
activities).

34
https://www.theice.com/publicdocs/ICE_Coffee_Brochure.pdf

33
 5. CREATING TEA FUTURES AND SWAP MARKETS: POTENTIAL CHALLENGES,
AND POSSIBLE WAYS TO ADDRESS THEM
Assuming that a tea futures market is difficult to establish, but possible with the support of the right
industry partners, this chapter will describe the concrete actions that would be necessary. Also, the
practicalities of structuring a contract around one “dominant” variety/location, or around an index, will
be discussed. Then, the possibilities to strengthen the chances of success for tea swaps will be debated.
Finally, the issue of speculation will be addressed.

Institutional arrangements
A commodity futures exchange is a relatively expensive bit of market infrastructure. Once in place, it can
be used to trade any kind of commodity, currency or other financial asset. The exchange is at the center
of a wider infrastructure, linking with brokers, banks, price reporting agencies etc. It does not cost an
existing exchange much to add a tea futures contract to its product range, or for brokers, to add tea
futures to their offerings to their clients. On the other hand, it would be expensive to develop this
infrastructure just for tea, and demanding in terms of specialist staff. Therefore, it is advisable to tie up
with an existing exchange for the creation of a tea futures market. The tea community could own the
intellectual rights to the tea contract, but not have an ownership stake in the futures exchange.

This is doable in China and India, where there are several competing exchanges. Kenya is in the process
of setting up a commodity exchange, which should be able to support the launch of a tea futures
contract (but cannot offer the immediate access to a large network of investors, traders, financiers etc.
that the China and India exchanges can offer). In Sri Lanka there is no commodity futures exchange; if
the tea industry there is interested in a tea futures contract (which as noted in the previous chapter
would address a rather small market – probably too small), it would be advisable to use the trading and
clearing platform of an international exchange (e.g., Dubai, India).

Ensuring convergence of the physical and futures market: physical

delivery, or financial settlement on the basis of an index?
Commodity futures markets have to reflect market conditions of the underlying physical commodities;
otherwise, they are just an inferior form of casino. To ensure such convergence of physical and futures
market prices, even though futures contracts normally do not lead to physical delivery, exchanges rely
on the final settlement mechanisms of their contracts.
Final settlement is the process of closing a futures contract as it approaches or reaches its expiry date.
Futures contracts are normally offered for a number of months (e.g., January, March, May, September,
December), and each contract has a last trading day. Final settlement can be in two forms. On the last
trading day, all open positions can be closed out, using a reference price to determine each contract’s
value. This is called financial settlement. Or alternatively, as the contract approaches delivery, one or
two weeks prior to expiry the contract’s delivery period starts, during which those who have sold
contracts can close out their position by physically delivering commodities as per the exchange’s

34
delivery specifications. While most commodity contracts have physical delivery, there are also many
contracts that rely on financial settlement.
When relying on financial settlement, the key is to choose a settlement price that is widely accepted by
the industry as providing a fair, unbiased and difficult-to-manipulate measure of the actual market value
of the commodity. Given the wide range of tea varieties and grades, no single variety/grade could be
considered representative, so in the case of tea financial settlement would have to be based on an index
of a range of varieties/grades. In the case of black tea, such an index can be constructed on the basis of
auction prices, applying econometric analysis to granular individual sales data. Teas that show price
behavior that is much different from that of other teas (e.g., strong seasonal variations) have to be
included from the index.
Alternatively, the contract’s settlement is through physical delivery. This implies delivery into a
warehouse; grading to ensure that the tea that is delivered meets the exchange requirements; the
application of discounts and premiums to delivered teas that are considered fair by the majority of
market participants; a system for assigning teas to the “long side” of the futures contracts (i.e., those
who had earlier bought futures contracts which will be extinguished through the delivery of physical tea)
that minimizes “pain” for these “long” parties; and a system that allows these parties to easily deliver
teas that they do now wish back to the exchange. All these conditions are in place in the coffee futures
market, but tea market participants may consider them daunting. In fact, the first three could be
handled through collaboration with the auction centers, the latter two by affecting delivery through an
electronic warehouse receipt system.
In a physical delivery system, the exchange has to list a range of varieties/grades/qualities that are
acceptable for delivery. Not all of the teas existing in the market, just a large enough range to make it
unlikely that any particular party can take control over most deliverable supply. These should be
grouped in baskets, and specific discounts and premiums should be established (valid at least for e year)
for each basket. Variability within the basket can be dealt with through a tasting process, as is done for
coffee futures – additional discounts/premiums can be imposed after delivery as a result of the tasting.
There should also be premiums and discounts for different delivery centers. The list of grades/varieties
that is deliverable against a futures contract should be specific for each contract month. 35
So, both financial and physical settlement appear doable, at least for black tea contracts (for green tea,
it may take a long time to construct a reliable mechanism to gather price data). Financial settlement is
easy for users and for the futures exchange, and avoids disputes on the quality of deliveries. But it does
not provide much certainty with respect to the price relation between one’s specific tea and the futures
price. Physical settlement poses a heavy work burden on an exchange, but it is easy for sellers as long as
they live close to an exchange delivery point. Also, sellers will know that if they sold a futures contract at
price X, that is the price they will actually realize, plus or minus the discounts/premiums established by
the exchange (which are more predictable than absolute price levels). On the other hand, physical
delivery may be cumbersome for buyers because they will receive at times varieties of tea that they do
not want, and for this reason, it is advisable that an electronic warehouse receipt system is used to

35
FAO, 2012a suggests correctly that in months where “pronounced upward or downward price variance for the
particular basket/grade is known to occur”, that particular basket/grade should be excluded from the deliverable
teas.

35
affect delivery.36 A similar risk of manipulation exists irrespective of whether a contract is physically or
financially settled – exchanges have countermeasures for both situations.

Building liquidity
A contract that is useful for a wide range of market participants is not automatically successful. Market
participants who say they are going to use the contract often do no not, giving as their reason that they
are waiting for the contract to become liquid (i.e., actively traded). But if everyone waits, of course the
contract will never take off. To remedy this common problem for futures exchanges, they collaborate in
the early years of the contract with one or more so-called market makers, with whom they share the
trading revenue. The market maker(s) agree(s) to provide, during part of the trading hours, a continuous
bid-ask spread (that is to say, they offer a price at which they are willing to sell, and a price at which they
are willing to buy); the bid-ask price levels will normally change continuously during the day.

Having a market maker is critical to achieve the initial growth of a new futures contract. When the
contract is related to an already-existing contract (e.g., an exchange introduces a crude oil contract in
India which is correlated to an already-liquid contract in the United States) it is straightforward to find a
market maker – the market maker just has to manage his market making risks on the US market, and
many companies have the systems and skills to do this. But for a completely new futures contract,
uncorrelated with any existing contract, the situation is different.

In fact, the best placed to provide a market-making role in such a new contract are major trading and
processing companies, who in the course of their ordinary business already take risk on both sides of the
market. Such companies can incorporate the market making risks they take into their overall risk
management practices, managing their market-making, buying and selling as parts of the same portfolio.
To make a tea futures contract viable, one or more significant tea companies have to be willing to play
this role. This will require commitment from such companies, and their belief in the usefulness and the
chance for success of the new contract.

Should the tea sector fear speculators?

The idea that speculators would and should become the largest group of participants in a tea futures
market can be discomforting. “Speculators” has a negative connotation in many languages. In fact, in a
futures market, “speculators” is a catch-all term for all market participants who are not active in the
physical tea market. They include:
- “Genuine” speculators, who on the basis of their analysis of the market feel that the futures
market is mispriced and will invest in such a way that, once the market returns to the “right
price”, they will benefit.

36
Exchanges normally arrange delivery through warehouse receipts: a seller deposits goods in an accredited
warehouse, where they are sampled and tested; then, a warehouse receipt is issues. To make delivery against a
futures contract, the seller then tenders this warehouse receipt to the exchange’s clearing house. The clearing
house transfers this warehouse receipt to a buyer, possibly in a way that tries to match what is delivered to the
buyer’s preferences in terms of location and quality. But in a paper system, all these steps are slow, and there are
also risks in the physical transfer of the paper receipts from one location to another. An electronic system is much
faster and cheaper to operate.

36
 - Portfolio investors, who have determined that by spreading their investments over a broad
range of non-correlated commodities they can improve their returns and/or reduce their risks.
- Gamblers (also called “noise traders”): people who invest in a commodity contract without
much understanding of the underlying commodity, but hope they will be lucky and win. While
they do not contribute information to the market, they do provide profit opportunities to other
speculators and hedgers.
- Arbitrageurs, who feel that a particular contract is mispriced in relation to another contract and
will take positions on both markets that result in a gain once the contracts have returned to
normal.
Missing from this list are “manipulators” – which is how many laypersons think about speculators.
Manipulators try to drive prices away from their fair equilibrium, while speculators help prices move
towards their equilibrium (“equilibrium” denotes the prices at which the market clears, i.e., supply and
demand match; in the short to medium run, equilibrium prices thus have little to do with production
costs or a “fair price” for the producer) . In fact, in order to manipulate a commodity futures contract,
one has to build up a dominant position in both the physical and the futures market. Speculators are not
active in the physical market and are thus in no position to engage in manipulation. But when they
notice that prices become misaligned because of a manipulation attempts, they will often trade against
the manipulator, helping to defeat his efforts.
Speculators try to take positions that become profitable when markets move back to a proper
supply/demand equilibrium; and through their actions (for example, buying when the product is “too
cheap”) they speed up this return of prices to their equilibrium. The fear that speculators could
somehow overwhelm the physical market and drive prices down is misplaced: in the case of a physically-
delivered contract, speculators who try to drive prices away from their equilibrium will end up with
stocks of the commodity that they can only sell at a large loss
Individually, speculators are therefore useful to markets. But in large numbers, they can have harmful
effects. For example, imagine the case of a portfolio investor who has half of his investments in crude
oil, and the rest split out over metals and agricultural commodities, including tea. Say that crude oil
prices double while other prices remain stable. Now, the share of crude oil in his portfolio is 66.7 per
cent, and in order to rebalance his portfolio he has to sell oil and buy other commodities, including tea.
In other words, price developments on one market will spill over into others, giving rise to sell or buy
pressure that has nothing to do with the actual supply/demand situation of the commodity. As investors
control vast sums of money, in particular compared to the relatively small sizes of many commodity
markets, their actions will at times disrupt the futures market and lead to mis-pricing. However, these
effects tend to be short-lived (a few days, because of possibilities to arbitrage between the physical and
futures markets) and give profit opportunities to hedgers; also, hedgers can reduce their exposure to
eventual price risks by somewhat adapting their hedging methods.
In all, the positive effects of speculation in creating liquid markets by far outweigh the occasional
negative effects on price formation.

37
 Tea swaps, a viable alternative?
While work towards tea futures has stalled, a different way to deal with tea price risk has been brought
to the market, by Teaswap (teaswap.co.uk). A trial started in March 2018 in Sri Lanka. 37 In future, the
company aims to also cover the Kolkata and Mombasa auctions. Teaswap acts as a broker for swaps.
Swaps are financial contracts, in this case for the exchange of differences in price between one auction
price reference and the other. Say market participant A wants to continue buying on the Mombasa
market, but wishes to eliminate his price risk. He can enter into a swap with Teaswap under which he
will receive the average of the Mombasa price for that period for an agreed period, but in exchange will
pay Teaswap a fixed price. To manage its price risk, Teaswap has to find a company with the opposite
exposure, which is willing to pay the Mombasa tea price and receive a fixed price. The structure is
illustrated in Figure 6.
Figure 10: managing price risk exposure through a tea swap
Mombasa Tea
Participant A (blender) pays Mombasa price for 100 tons
Auction

 A pays broker  Broker pays Mombasa

a fixed price price for 100 tons
for 100 tons

Swap broker

 Broker pays B  B pays broker

a fixed price Mombasa price
for 100 tons for 100 tons

Mombasa Tea
Participant B (factory) receives Mombasa price for 100 tons
Auction

As long as the fixed price Teaswap receives is higher than what it pays, the company is profitable.
However, it runs the risk that either A or B will default on its swap, leaving Teaswap with the obligation
to pay one party without receiving anything from the other..
A brokerage function can be useful. In principle, A and B could have entered into a direct fixed price
forward contract with each other that would have the same effect. But for this, they first need to know
of each other’s interest in such a contract; and then they would have to be willing to take the credit risk
on the other party. A broker can fill this information gap.
Swaps like this are common in energy and metals markets, and also exist in agricultural markets.
However, developing such a product in the tea market has its difficulties, which relate to the ability of
Teaswap to manage its own risk. First, Teaswap will have to manage counterparty risk – the risk that
one of its counterparties defaults on its obligations. This is not easy. Not many players on the tea market
have a high credit rating. If Teaswap accepts partners with a lower credit rating, it will either have to

37
http://www.ft.lk/agriculture/Tea-Swap-trading-begins-in-Sri-Lanka/31-651485

38
absorb the risk (putting capital aside against the risk of default, just like banks do with risky loans), or
require the partners to put up collateral such as bank guarantees (which are expensive). It should be
noted that if the counterparty were a factory that hedges its “made tea” price risks for tea bought from
smallholders, in case prices rise this factory is very unlikely to be able to convince smallholders to
continue selling to it when market prices increase. Thus, if it is to continue operating, it has to pay its
suppliers on the basis of the going market price, while having locked in a lower price itself – it will have
to absorb this risk. Even if the counterparty is a cooperative, its members can be expected to protest
vociferously against the idea of delivering at a below-market price, and are likely to force the
cooperative to default on its obligations to the broker.
Second, it is very unlikely that any market can be developed if Teaswap does not take any price risk –
and it is reported that it does not intend to do so. 38 Teaswap aims to only match opposing parties (as in
the figure above). However, this is not how swap markets in other commodities were able to develop.
Particularly when the swap market is still under development, the likelihood that Teaswap will get
opposite interests at around the same time is rather small (even if its staff actively market to the tea
companies).39 Just as has been the case of other commodities, if it wishes to expand the swaps market,
Teaswap has to take one-sided risks (price risk for which there is no direct commercial counterparty). So,
to grow the market Teaswap has to be willing to accept tea price changes.
This is quite common in energy and metals markets, and brokers manage this risk through a
combination of sophisticated financial modeling, temporarily “warehousing” positions on futures
exchanges, and in many cases, engaging in physical trade. But in the absence of a futures market, this
becomes difficult, albeit not entirely impossible – this is further discussed below.

Improving the chances of success for tea swaps

Tea swaps face two major constraints:

- The intermediary broker has to take on counterparty risk on both sides of the transaction. If one
side defaults, the broker is left with payment obligations to the other side. To avoid this risk, the
broker is likely to deal only with highly creditworthy counterparties, or ask for credit
enhancement (in the form of security deposits or bank guarantees).
- The likelihood of two participants with an opposite interest in the same market, same variety,
same timeframe, same quantity, coming to the swap broker at the same time is very small.

Commodity exchanges deal with the first problem by operating a clearinghouse. In futures trade, when a
bid by a buyer matches an offer by a seller, the clearinghouse immediately interposes itself between
buyer and seller: the buyer buys from the clearinghouse, the seller sells to the clearinghouse. Buyer and
seller do not have a credit risk to each other, but to the highly rated clearinghouse. The clearinghouse
manages its risks towards buyer and seller by having both of them pay a margin deposit, and then ask
38
Tea Swap trading begins in Sri Lanka, Daily FT, 19 March 2018.
39
In active futures markets there is no such problem as most of the trade is by speculators, who as a group usually
are active on both sides of the market. Thus, hedgers can always find a ready counterparty. Apparently some tea
industry representatives consider the absence of speculators from the tea swap market a reason to prefer it over
the futures market, but this reflects a misunderstanding of the role and functioning of speculators.

39
for replenishment of this margin on a daily basis to ensure that no one has an incentive to default on his
obligations. Because the margining process is daily, margins just need to be sufficient to cover price
fluctuations over a two-day time frame, compared to the much higher margins that a swap broker
would have to ask to cover price risks over, say, a one-month period.
To manage its counterparty risk, Teaswap could in principle collaborate with a clearinghouse.
Clearinghouses have the ability to clear swaps, as long as they have access to good price data. The swap
broker can outsource its counterparty risk management to the clearinghouse, making it possible for the
broker to enter into swaps with any interested party. For the counterparties, however, this means they
have to be willing to pay margin deposits (guarantees) which increase when the value of their swap
moves against them – note that these are payments that have to be made before the tea that is hedged
through the swap is actually sold, so the counterparty needs easy access to finance. Thus, even if
Teaswap could find a clearinghouse willing to manage the counterparty risks in the tea market, most tea
companies would not be able to meet the clearinghouse’s conditions.
Another way to reduce counterparty risk is through an auction. Teaswap and “tea-selling” clients could
both sign an agreement with the auction that is being used by the client, under which the auction
commits to prioritize Teaswap in the distribution of the funds received for the client’s sale of tea. In
other words, as long as the company continues selling tea through the auction, its commitments to
Teaswap will be met. However, scenarios in which the tea auction will withhold payments to a
plantation to pay the money instead to Teaswap will lead to strong tensions, and internal pressures to
find a way around the assignment of the payment (such as sale under the name of another plantation).
And it does not cover situations where plantations/factories can just decide to sell outside of the auction
(or farmers who had committed to sell to a factory for a fixed price can just sell elsewhere), nor does it
cover defaults by the buyer.
With respect to the second problem – to develop the market the broker will have to run price risk –, the
experience in other markets is that swap brokers can in fact warehouse positions, and manage the
resultant risks through sophisticated financial modeling, futures trading and physical trading. To
understand how this would work, consider that investors are willing to take any kind of risk, as long as
they are properly compensated for it. Thus, if a risk is AAA, they may want to receive a 4 percent
interest rate on their investment; if it is BBB, they may need 10 percent; for CCC risk (“junk bonds”) they
may want an interest rate of 30%. If the swap broker has a series of positions – say, “sell” the price risk
on the Mombasa auction for 12 months; “buy” the Mombasa risk for half that quantity and for only 6
months; buy the Colombo price risk for the same quantity but for 18 months. These various positions
partly offset each other. With sufficient historic price data, it is possible to calculate with a reasonable
level of certainty what is the risk that the swap broker will lose money. Say that this risk is similar to the
risk of investments that are now qualified as “BBB”; then, if the swap broker can pay investors an
interest rate of 11%, they will be happy to cover that risk (as long as the investment is not too small; in
the initial stages of the swaps market, the industry may have to put up the necessary risk capital). The
swap broker, in turn, can make this return through the difference in prices that it charges buyers and
sellers…
The grade-specific data for proper financial modeling may not be publicly available in the tea market,
but the auctions and large tea companies may have much of these data in their private databases. If the

40
tea industry is keen on getting swaps to work, it should share all price data that it has with Teaswap (ad
other interested swap brokers), and encourage tea auctions to do the same.
Having a proper financial model will allow a broker to build up a portfolio of diverse positions –
guaranteeing to pay a fixed price for A tons of grade X, but receive a fixed price for B tons of grade Y.
They just have to pass on the risk premiums that come with such imperfect hedges to the buyers and
sellers; and assuming that the swap broker is able to build up a sufficiently large portfolio, he can lay off
much of the risks with outside investors. But while this may work as long as the commercial companies
are willing to pay, this is still a somewhat risky process; it would be much easier for a swap broker if he
could lay off “generic” market risk to a tea futures exchange (leaving him with the risk of varying
premiums/discounts for specific grades).
In all, the prospects of the tea swap market, in the absence of a futures market, do not seem bright. In
its current format, the proposal has significant drawbacks (counterparty risk, the need to exactly match
opposing interests) that make it impossible for the swap market to develop beyond a small fraction of
the tea industry. There are ways to improve on the current proposal: first, using the auctions to enforce
the swap obligations of tea-selling companies, and second, developing an econometric model through
which investors can be enticed into the tea swap market thus permitting “mismatched” buy- and sell-
swaps. However, even with these mitigants in place, most tea companies will still fall outside of the
scope of the swaps market. Moreover, to put these mitigants in place is complex and needs active
industry collaboration. If the tea industry is willing to commit to such enthusiastic support for tea swaps,
it may consider that with a lesser effort it could instead develop tea futures contracts, and that these
contracts can be used by all of the tea industry.

41
 6. BUILDING INCLUSIVE MARKETS: HOW TO BRING GROWERS ON BOARD
Typically, up to now growers do not directly use futures markets. The main reason is that futures
positions have to be managed on a daily basis, including through margin payments that, if demanded by
the exchange’s clearing house, have to be made within a working day. Growers typically do not have the
time for this, not the easy access to money which daily margins would require. However, the spread of
mobile phones and electronic money are changing this, and some growers may nowadays find it
possible to trade directly. A new tea exchange may wish to build on these trends and build direct
connectivity to digitally-empowered growers.
Rather, growers have been benefiting from futures markets because these provide them with superior
market information; and they can indirectly use futures exchanges through cooperatives, processors and
traders.40 In the experience of the Indian commodity exchanges (which became active in the first half of
the 2000s), farmers shifted their usual time to go to the market to sell their produce after start of
futures contracts: instead of going early in the morning, they now went after 9:30, when the futures
market opened and farmers could use the price information to better bargain with traders. Indirect use
of futures markets has also been prevalent – most of the time through processors, which give the
growers who supply them the possibility to lock in (minimum) forward prices. In a competitive
environment, traders will also use futures exchanges to offer an expanded range of pricing options to
growers.
It should not be assumed that growers automatically know how to use an exchange to improve their
risk/reward profile. Many growers may instead be tempted to speculate. For example, if they want to
lock in high tea prices which they see on a futures market, they may sell tea futures beyond their
capacity to actually supply physical tea. If prices subsequently increase further, they do not have
sufficient tea to deliver against all their short positions, and instead have to cover their losses financially.
Good education on what constitutes hedging and what is speculation is essential. Where growers
appoint an agent to engage in the futures market in their stead (such as a cooperative), good corporate
governance needs to be put in place to ensure that the cooperative does not speculate with the
growers’ money. Donor agencies and governments may wish to give technical assistance to this effect.
However, actions to move growers closer to the point of sales to blenders can have a bigger impact on
growers’ income then helping them negotiate a better leaf price with an itinerant purchasing agent. In
South India, for example, it was found that by selling tea through self-help groups, growers in Tamil
Nadu and Kerala could improve their leaf price from INR 4-9 to INR 10-15 per kg. 41

40
See for an overview with respect to coffee farmers Rutten, 2007.
41
Murray, 2013.

42
CONCLUDING REMARKS
It is, prima facie, surprising that there is a vibrant futures market for cocoa and coffee, but not for tea.
Futures markets have long since been found useful by all the participants in cocoa and coffee value
chains, so why would tea market participants deprive themselves of this support? One reason is that
there has been more pressure in the cocoa and coffee sectors because these crops are seasonal, with
higher risks of market disruptions and a greater need for storage (and thus, exposure to the risk of price
falls for those owning the stored produce). Another reason is that the tea sector seems slightly more
complicated, with a market that is more segmented than the cocoa and coffee markets.
Nevertheless, the differences between tea, on the one hand, and cocoa and coffee on the other are not
radical, but a matter of degree. Cocoa and coffee markets are segmented, tea a bit more so. Cocoa and
coffee have many grades and varieties, tea a bit more so. Ultimately, for many beans, the quality of
cocoa and coffee can only be determined by actually tasting the product; perhaps for tea, this is true for
a larger segment of the market. Therefore, it would seem that the practices adopted in, say, the coffee
sector, to match the need for standardized futures contracts with the reality of a highly diverse physical
product, can be used in the tea sector too.
Futures markets are in continuous search of new products. They will, however, go for the easier
products first. Tea does not make that list. Therefore, the initiative to introduce tea futures has to come
from the tea sector itself. They would have to engage established futures markets to obtain relevant
expertise, gain access to trading and clearing platforms etc., but without such a drive by physical market
players the futures industry is not likely to move ahead.
Much further work is still required to establish whether it is worthwhile to invest in the creation and
promotion of a futures market. In China, this starts from the basic groundwork of building a knowledge
base on tea trading and market practices; in India, it consolidating earlier work on varieties/grades to
establish which exact teas can form the basis for a standardized futures contract.
But the starting point of this report should be kept in mind: “Tea producers and exporters are plagued
by repeated cycles of price and revenue fluctuations, endangering the sustainability of a significant part
of the industry.” Futures contracts are not a panacea to all of the producers’ and exporters’ problems,
but they are effective, proven, low-cost solutions for dealing with price and revenue fluctuations. It thus
makes sense for the tea industry to act as champion of tea futures contracts, and formulate action plans
for each of the relevant tea market segments.
One action that is underway is the development of tea swaps. These are bespoke arrangements,
wherein market participants exchange their exposure to variable market prices against fixed prices that
they either receive from a broker (if they are a tea seller), or pay to the broker (if they are a tea buyer).
The issue, however, is that the broker will have to take on considerable counterparty and market risk if
the swap market is to develop – it is unlikely that a seller’s interest in receiving a fixed price for a specific
quality and quantity of tea is matched, at any given time, by a buyer’s interest in paying a fixed price for
that same quality and quantity. There are ways for a swap broker to enhance its capacity to take on risk
(in particular, by collaborating with auctions of with a clearinghouse), but the tea swap market has its
limits. It cannot replace the broad range of services and ready accessibility of a tea futures market.

43
REFERENCES
Brouwer (2014), Ann-Marie, Simon Billing and Sally Uren, The future of tea – a hero crop for 2030,
Forum for the future
Chang, Kaison (2015), World tea production and trade. Current and future development. FAO
Intergovernmental Group on Tea, FAO
FAO (n.d.), Smallholder Tea Producers Benefit from Harmonized Safety Standards
<http://www.fao.org/in-action/smallholder-tea-producers-benefit-from-harmonize-safety-
standards/en/>
FAO (2003), Examination of the transmission of world tea prices, Intergovernmental Group on Tea, 15th
Session, 20-22 July, CCP:TE 03/5
FAO (2005), Application of price transmission on selected tea markets, Intergovernmental Group on Tea,
16th Session, 20-22 July, CCP:TE 05/5
FAO (2012a), Prabhat Bezboruah, The future of futures contracts in tea: an Indian scenario,
Intergovernmental Group on Tea, 20th Session, 30 January-1 February, CCP:TE 12/CRS 3
FAO (2012b), Risk management practices in tea markets: a case study of possible futures contracts in
India, Intergovernmental Group on Tea, 20th Session, 30 January-1 February, CCP:TE 12/3
FAO (2012c), Risk management through introduction of futures contracts in tea - an attempt at solving
the heterogeneity problem, presentation to the Intergovernmental Group on Tea, 20 th Session, 30
January-1 February
FAO (2014), Market report - China, Intergovernmental Group on Tea, 21st Session, 5-7 November, CCP:TE
14/CRS4
FAO (2017), Report of the 71st Session of the Committee on Commodity Problems (Rome, 4–6 October
2016), C 2017/22
FAO (2018), Emerging trends in tea consumption: informing a generic promotion process.
Intergovernmental Group on Tea, 23st Session,17-20 May, CCP:TE 18/2
Gori, F. (2017), Unique tea qualities that determine auction prices
<https://www.businessdailyafrica.com/analysis/tea-qualities-that-determine-prices/539548-3884858-
6uvfpd/index.html>
Goswami, R. (2017), Tea Board promotes auctions - Proposal for all produce
<https://www.telegraphindia.com/1170622/jsp/northeast/story_158104.jsp>
Groosman, Michael (2011), Sector overview – tea, IDH – the sustainable trade initiative, October
Hall (2000), Nick, The Tea Industry, Woodhead Publishing Limited
Hazarika, K.(n.d.), Tea Auction Market With a special reference to Guwahati Auction Centre

Herath, D. & Weersink, A. (2006), Structural Changes in the Sri Lankan Tea Industry: Family Farms vs.
Plantations
Hossain, T., Khalil, F. Shum, M. (2016), Auctioneers as Market Makers: Managing Momentum in
Chittagong Tea Auctions

44
International Institute for Sustainable Development (IISD, 2014), Tea Market, Chapter 14 in State of
Sustainability Initiatives
Jacobs, A. (2009), A County In China Sees Its Fortunes in Tea Leaves Until a Bubble Bursts. The New York
Times <http://www.nytimes.com/2009/01/17/world/asia/17tea.html >
Kadavil, S. M. (2007), Indian Tea Research, SOMO research
Kenya Information Guide (n.d.), Kenya Tea: About the Black and Green Tea Grown in Kenya
<http://www.kenya-information-guide.com/kenya-tea.html >
Morser, Anna (2010), A bitter cup: The exploitation of tea workers in India and Kenya supplying British
supermarkets, War on Want, July
Murray, A. (2013) Smallholder Tea Producer Experiences of Voluntary Private Standards

Musuva, G., Lewa, P., Achoki, G. &Luciani, A. (2016), Value Chain Risk Analysis For Small Holder Tea
Farmers in Kiambu Country – Kenya, International Journal of Economics, Commerce and Management,
Vol. IV
Mutegi, N. M. (2013), An Empirical investigation Of Factors Influencing Tea Export Earnings in Kenya
1973-2013
Panhuysen, S. (2010) Tea Barometer 2010, Tropical Commodity Coalition
Rauhala, E. (2009),PuErh Tea: China's Next Hot Commodity?
<http://content.time.com/time/business/article/0,8599,1918419,00.html >
Rutten (2007), Lamon, and Frida Youssef, Market-based price risk management - An exploration of
commodity income stabilization options for coffee farmers, International Institute for Sustainable
development
Sarkar, K. (2013), Trends and Price Formation Mechanism in Indian Tea Auctions, National Research
Programme On Plantation Development (NRPPD), Discussion Paper No. 23
SOMO (2008), Sustainability Issues in the Tea Sector: A Comparative Analysis of Six Leading Producing
Countries
Tanui (2012), John Kipkorir, Wanping Fang, Weiying Feng, Peifen Zhuang & Xinghui Li, World Black Tea
Markets: Relationships and Implications for the Global Tea Industry, Journal of International Food &
Agribusiness Marketing Vol. 24, Issue 2
Teabox (2017), Orthodox VS CTC Tea<https://tea101.teabox.com/orthodox-vs-ctc-tea/>
Tea Class (n.d.), Types of Tea <https://www.teaclass.com/lesson_0102.html>
Teasenz (2017), The 2007 PuErh Tea Market Crash: The Boom And Bust Revisited
<https://www.teasenz.com/chinese-tea/2007-pu-erh-tea-market-crash-boom-bust-crisis-
recession.html#comments>
United Nations Conference on Trade and Development (UNCTAD, 2016), Tea – an INFOCOMM
Commodity Profile

45
Waireri (2016), Catherine Nyokabi, An exploration of the possible impact of commodity futures trading
for tea farmers in Limuru to mitigate the price volatility of tea, School of Finance and Applied Economics
Strathmore University, Nairobi, Kenya, November

46
Annex A - Frequency Distribution of Price Changes of the Indian and Sri Lanka Auctions

Table A1. Frequency Distribution of Price Table A2. Frequency Distribution of

Changes of Guwahati Average Weekly Price Changes of Siliguri Average
Auction Prices, 2008-2017 Weekly Auction Prices, 2013-2017

% Price
Frequency Frequency Frequency Frequency
% Price changes changes
Distribution Distribution Distribution Distribution
compared to compared
(Increasing (Decreasing (Increasing (Decreasing
previous month to previous
price) price) price) price)
month
0-5% 35% 48% 0-5% 32% 51%
5-10% 2% 5% 5-10% 3% 7%
10-15% 1% 2% 10-15% 0% 0%
15-20% 0% 0% 15-20% 1% 1%
20-25% 1% 0% 20-25% 1% 1%
25-30% 1% 0% 25-30% 0% 0%
30-50% 0% 1% 30-50% 0% 0%
>50% 1% 0% >50% 1% 0%

Table A3. Frequency Distribution of Price Table A4. Frequency Distribution of

Changes of Jalpaiguri Average Weekly Price Changes of Cochin Average
Auction Prices, 2008-2017 Weekly Auction Prices, 2013-2017
% Price
Frequency Frequency Frequency Frequency
% Price changes changes
Distribution Distribution Distribution Distribution
compared to compared
(Increasing (Decreasing (Increasing (Decreasing
previous month to previous
price) price) price) price)
month
0-5% 25% 37% 0-5% 45% 38%
5-10% 4% 15% 5-10% 4% 5%
10-15% 5% 8% 10-15% 1% 2%
15-20% 1% 3% 15-20% 1% 0%
20-25% 2% 0% 20-25% 0% 1%
25-30% 0% 0% 25-30% 0% 0%
30-50% 0% 0% 30-50% 1% 1%
>50% 0% 0% >50% 1% 0%

47
Table A5. Frequency Distribution of Price Table A6. Frequency Distribution of
Changes of Coonoor Average Weekly Auction Price Changes of Coimbatore Average
Prices, 2008-2017 Weekly Auction Prices, 2013-2017

% Price
Frequency Frequency Frequency Frequency
% Price changes changes
Distribution Distribution Distribution Distribution
compared to compared
(Increasing (Decreasing (Increasing (Decreasing
previous month to previous
price) price) price) price)
month
0-5% 45% 43% 0-5% 46% 40%
5-10% 4% 4% 5-10% 6% 4%
10-15% 1% 0% 10-15% 0% 1%
15-20% 1% 1% 15-20% 0% 0%
20-25% 0% 0% 20-25% 0% 0%
25-30% 0% 0% 25-30% 0% 0%
30-50% 0% 0% 30-50% 0% 0%
>50% 0% 0% >50% 0% 0%

Table A7. Frequency Distribution of Price

Changes of Colombo Average Monthly
Auction Prices, 2014-2017
Frequency Frequency
% Price changes
Distribution Distribution
compared to
(Increasing (Decreasing
previous month
price) price)

0-2% 11% 23%

2-4% 13% 21%
4-6% 11% 11%
6-8% 6% 0
8-10% 2% 0
>10% 2% 0

48
Annex B. Scatter Graphs Indicating the Price Correlations between Auctions

B1 B2

B3 B4

B5 B6

49
50