Economic Policy and Cotton in Uzbekistan

Abstract

Uzbekistan is the seventh largest global cotton producer and third largest cotton supplier for world markets. Uzbekistan's Government policies largely shield cotton producers from world market price signals, and cotton area has changed little over the past decade despite strong international price fl uctuations. Government pricing and exchange rate policies tax cotton producers and more than offset the value of input subsidies for cotton growers. The degree of taxation declined for several years after 2000, but increased again in the late 2000s. In the 2009, cotton output dropped as Uzbekistan responded to reduced water availability and increased global food prices with higher taxes on cotton growers. With continued taxation of cotton production, Uzbekistan likely will continue to lose ground to more dynamic cotton exporters like India and Brazil. As a result, USDA's longrun baseline projections for Central Asia show that the region's share of world cotton production will continue to fall over the next decade.

Full text

Abstract
Uzbekistan is the seventh largest global cotton producer and third largest cotton
supplier for world markets. Uzbekistan’s Government policies largely shield cotton
producers from world market price signals, and cotton area has changed little over the
past decade despite strong international price fl uctuations. Government pricing and
exchange rate policies tax cotton producers and more than offset the value of input
subsidies for cotton growers. The degree of taxation declined for several years after
2000, but increased again in the late 2000s. In the 2009, cotton output dropped as
Uzbekistan responded to reduced water availability and increased global food prices
with higher taxes on cotton growers. With continued taxation of cotton production,
Uzbekistan likely will continue to lose ground to more dynamic cotton exporters like
India and Brazil. As a result, USDA’s longrun baseline projections for Central Asia
show that the region’s share of world cotton production will continue to fall over the
next decade.
Acknowledgments
The author acknowledges the assistance of Armelle Gruere of the International
Cotton Advisory Committee in developing the Uzbek price database and the assis-
tance of TCX Investment Management Company B.V. in acquiring exchange rate
data. Erik Dohlman, Mark Jekanowski, Maurice Landes, and Leslie Meyer from
USDA’s Economic Research Service; Mark Lindeman, Clay Hamilton, and Nizam
Yuldashbaev from USDA’s Foreign Agricultural Service; Richard Pomfret from The
University of Adelaide; Carol Skelly from USDA’s Offi ce of the Chief Economist;
and additional anonymous reviewers provided helpful comments and insights. This
study benefi tted signifi cantly from the attention and knowledge of these reviewers,
but any remaining errors are the responsibility of the author. The author also
acknowledges and appreciates the editorial assistance of Angela Anderson and the
design assistance of Wynnice Pointer-Napper of ERS.
United States
Department
of Agriculture
www.ers.usda.gov
A Report from the Economic Research Service
Stephen MacDonald, stephenm@ers.usda.gov
Economic Policy and Cotton
in Uzbekistan
CWS-12h-01
October 2012
Contents
Approved by USDA’s
World Agricultural
Outlook Board
Introduction . . . . . . . . . . . . . . . . . 2
Uzbekist an’s “State Or der”
System and Cot ton . . . . . . . . . . 4
Uzbek Cotton Pricin g . . . . . . . . . . 5
Exchange Rate Policy . . . . . . . . . . 7
Calculat ing the Exp ort Parity
of Domestic Prices . . . . . . . . . . 9
The Role of Market ing Cost s . . 11
Nominal R ates of Ass istance ,
by Policy . . . . . . . . . . . . . . . . . . 12
Subsidies: Exte nsive and Hard
To Measu re . . . . . . . . . . . . . . . 13
Taxation Of fsets Subs idies for
Uzbek Cotton. . . . . . . . . . . . . . 14
Linkin g Policies To Cot ton
Production . . . . . . . . . . . . . . . . 15
Other Infl uences on Yield . . . . . 17
Factors Driv ing Change s
in Net Taxation . . . . . . . . . . . . 18
Conclusions . . . . . . . . . . . . . . . . . 20
References . . . . . . . . . . . . . . . . . . 21
Appendix: Nominal Rate of
Assistance (NRA) . . . . . . . . . . 25

2
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
Uzbekistan is one of the world’s largest cotton exporters.1 Central Asia2
accounts for 16 percent of world cotton trade, and Uzbekistan remains the
leading producer and exporter among these former members of the Soviet
Union. Uzbekistan accounted for 25 percent of world cotton trade in the
1970s and 1980s, more than any other exporter. At that time, its per hectare
cotton yield was consistently among the highest in the world, exceeding U.S.
yields by 74 percent during the fi rst half of the 1970s (fi g. 1).
Since then, Uzbekistan’s cotton yield has trended downward, and the Uzbek
share of world trade has slipped below 10 percent (fi g. 2). Elsewhere around
the world, yields have increased, with annual gains among major exporting
countries ranging between 1.6 percent (United States) and 4.3 percent
(Brazil). Uzbekistan has successfully transitioned into an Asian supplier as
its traditional markets in Russia and Ukraine have dwindled, but with output
now about half of its 1985 level, it may not be meeting its full potential.
It is questionable whether Uzbekistan could sustain a return to its mid-1980s
production levels or whether it would want to. Soviet planners oversaw an
enormous increase in Central Asia’s irrigated cotton area between 1945 and
the 1980s. The consequent streamfl ow reductions and runoff contamination
by fertilizers and other chemicals substantially reduced the area formerly
covered by the Aral Sea, resulted in toxic residue along former coastal
regions, and reduced crop yields in some provinces. In addition, wide-
spread reports of child labor in Uzbekistan’s cotton harvest add yet another
dimension to the evaluation of the country’s optimal cotton output levels
(Environmental Justice Foundation, 2005).3 Recent weather problems and
disputes with neighboring countries over irrigation supplies suggest that,
even with a return to mechanization, Uzbekistan’s current maximum produc-
tion potential may be lower than it once was. Regardless of Uzbekistan’s
potential for sustainable production, a review of the country’s economic and
agricultural policies shows signifi cant government intervention that currently
hinders cotton production, reducing output.
1All data on agricultural output,
consumption, and trade came from
USDA’s Foreign Agricultural Service
Production, Supply and Distribution
(PS&D) database for global agricul-
ture, unless specifi cally noted other-
wise. Units of measurement follow
USDA conventions for international
data, which include stating cotton data
in fi ber terms rather than as seedcotton,
even in reference to countries where
farmers market the latter.
2Central Asia refers to the former
Soviet Republics of Kazakhstan,
Kyrgyzstan, Tajikistan, Turkmenistan,
and Uzbekistan.
3Since it gained independence,
Uzbekistan has largely abandoned the
formerly widespread use of mechanical
harvesters (Pomfret, 2000), possibly
as a result of the disruption of critical
supply chains with the breakup of the
Soviet Union. Financial distress on
Uzbek cotton farms also may have
reduced mechanization (Isengildina et
al., 1998). The USDA attaché reports
that wages for cotton picking are sub-
stantially higher in Kazakhstan and the
Kyrgyz Republic than in Uzbekistan
(USDA/FAS, 2005).
Introduction
Figure 1
Uzbekistan’s cotton yields lag
Kilograms per hectare
Source: USDA, Foreign Agricultural Service, Production, Supply and Distribution Online data,
2012; USDA, Foreign Agricultural Service, 1977.
1970 75 80 85 90 95 2000 05 10
0
500
1,000
1,500
2,000
2,500 Australia China United States Uzbekistan India

3
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
This study estimates the degree of taxation implicit in the policies used by
Uzbekistan to set the domestic producer price of cotton, and calculates the
impact of these policies in terms of a Nominal Rate of Assistance (NRA)
(see box, “Nominal Rate of Assistance”). We combine information about
Uzbekistan’s cotton sector from several recent studies, assess the net degree
of direct and indirect taxation of cotton in Uzbekistan, and examine how this
taxation has varied over time.
Figure 2
Uzbekistan cotton production and share of world trade, 1970-2012
1,000 bales (production)
Source: USDA, Foreign Agricultural Service, Production, Supply and Distribution Online data,
2012; USDA, Foreign Agricultural Service, 1977.
1970 75 80 85 90 95 2000 05 10
0
2,000
4,000
6,000
8,000
10,000
0
5
10
15
20
25
30
35
Production
Share of world trade
Percent (share of world trade)
Marketing year
Measuring agricultural support for a particular commodity requires measuring
direct subsidies provided to or taxes imposed upon producers, but also analysis
of:
• Relevant trade barriers;
• Support to sectors providing inputs; and
• Whether domestic consumption is subsidized or taxed.
The presence of multiple exchange rates also must be accounted for in countries
that enforce capital controls. Josling et al. (2010) summarized the various
accounting conventions used to measure agricultural support programs. This
study follows the approach described in Anderson et al. (2009) and calculated
a nominal rate of assistance (NRA). We used data on the export parity of
Uzbekistan’s procurement prices—differentiated by offi cial and parallel
exchange rates—to estimate the level of implicit and explicit producer taxation
(see Appendix for details).
Nominal Rate of Assistance

4
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
Since its independence from the Soviet Union in 1991, Uzbekistan has
followed a strategy of gradual transition from planned to market economy.
The government continues to exercise extensive control in agriculture, partic-
ularly in cotton and wheat production, which are referred to as “centralized”
crops. The State maintains ownership of the land, and the right to use land for
agriculture (other than household plots) is conditional on acceptance of the
State’s quotas for planting cotton and wheat. The State also provides subsi-
dized inputs, including irrigation.
The “State order” system also includes quotas on the production of cotton and
wheat, as well as on area planted. To ensure quotas are met, the State moni-
tors efforts year-round (Veldwisch and Spoor, 2008): leaching4 is monitored
in the winter; planting area, varieties, and dates are determined by the State
in the spring; and fertilizer application during the growing season is directed
by Ministry of Agriculture and Water Resources (MAWR) offi cials. During
the cotton-growing season, State offi cials visit farms to determine yield
potential and adjust planning targets and production quotas. Annual planting
area is determined by a State plan and, at the local level, planning may
extend to determining which fi elds are used for cotton, wheat, or noncentral-
ized crops. While the State and collective farms organized during the Soviet
period have been largely privatized since 2006, a large number of farms still
rely on centralized Machine Tractor Parks (MTP) for machinery, and MTPs
prioritize centralized crops.5
The textile industry exemplifi es Uzbekistan’s partial transition from planned
to market economy. Uzbek cotton consumption has grown in recent years,
representing greater foreign investment and an effort by the government
to increase the share of fi ber processed locally rather than exported. Many
textile enterprises are joint ventures, with the government as the main share-
holder (Rudenko, 2008). The Ministry of Textiles has been reorganized into
the State Joint Stock Company (SJSC) UzbekEngilSanoat, which manages
the government’s shares in textile fi rms and is also engaged in promoting the
export and domestic sales of local textile output. SJSC UzbekEngilSanoat
also promotes investment in the textile industry (Naumov et al., 2010).
4Irrigation can result in soil sali-
nization. Extra water, in addition to
crop requirements, can be applied
to irrigated fi elds to leach these salts
away (Ayers and Westcot, 1994). In
Uzbekistan, this commonly occurs
during winter.
5MTPs were formerly a component
of the State’s authority over agricul-
tural producers, but the relationship has
evolved into what one study describes
as a “patron-client” relationship dis-
tinct from formal authority (Shtaltovna
et al., 2011).
Uzbekistan’s “State Order” System
and Cotton

5
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
The main policy instrument that transfers resources from cotton production
is a procurement price set below world prices. Procurement below world
prices is possible because producers are effectively required to sell their
cotton exclusively through offi cial channels. While much of Uzbekistan’s
cotton is exported, international trade is strictly controlled. This restriction
limits global price transmission and allows resources to be transferred from
producers, either directly by the government or indirectly through semi-
governmental fi rms.
International cotton prices adjust relatively freely, so exporting countries
seldom sustain signifi cant gaps between their export prices and the world
price. Uzbekistan regulates its export price, but does so in a relatively trans-
parent manner. The Uzbek foreign trade companies (FTCs) authorized to
export under the Ministry for Foreign Economic Relations, Investments, and
Trade6 price cotton based on the A Index7 and Cotlook’s quote for Uzbek
cotton. Exceptions occur in some years, but they are diffi cult to sustain.
For example, Uzbekistan’s Government temporarily imposed a minimum
export sales price of 60 cents per pound in December 2008, (Cotton Outlook,
2009). World prices averaged well below that level for several months during
2008/09, resulting in an unprecedented near-doubling of ending stocks, to 50
percent of use.
Policymakers in Uzbekistan have to take the world price of cotton as given
when determining export prices, but the determinants of the State procure-
ment price (SPP) are unclear. Djanibekov et al. (2010) suggested that the
SPP is established annually based on the world price minus marketing
costs. Guadagni et al. (2005) describe such a mechanism as well, but
noted that costs were overstated. Sadler (2006) reported that prices were
calculated based on production costs adjusted by recent infl ation and
asserted that, in earlier years, prices were fi xed as a percentage of interna-
tional prices, ranging from 70 percent in 1996 to 85 percent in 1997 and
100 percent in 2000. Rudenko et al. (2009) describe a process where the
semi-governmental State joint stock company that monopolizes ginning
in Uzbekistan, UzPakhtaSanoat, negotiates a price with the FTCs. For our
purposes, this last observation is perhaps the most useful, highlighting the
fact that price is determined by balancing competing interests within the
Uzbek Government.
Volatile world prices hinder policymakers’ ability to precisely target the rela-
tionship between the SPP and the world price; they have to choose between
recent price movements and longer-term averages when forecasting the world
price. One study (Sadler, 2006) noted that, at least prior to 2005, producers
entered into a sales contract with their local gin, and that, theoretically, the
price would be adjusted if the market price changed. In practice, the price
was set so low that even a decline in the market price left it above the contract
price. Thus, the system acted as a price smoothing mechanism, reducing SPP
volatility compared with the world price (fi g. 3). Rudenko et al. (2009) noted
that unforeseen price changes between the time the prices are initially negoti-
ated and the time cotton is fi nally exported are absorbed by the FTCs, effec-
tively smoothing farm prices. Pomfret (2009) also characterized the State
6Three FTCs account for virtually
all of Uzbekistan’s cotton exports:
Joint-Stock Company Uzinterimpex,
State Joint Stock Foreign Trade
Company Uzmarkazimpex, and
Uzprommashimpeks. About 75-80
percent of Uzbekistan’s cotton produc-
tion is exported. Cotton for domestic
mill use is purchased through the
Uzbekistan Commodity Exchange,
which also accounts for a portion of
cotton exports.
7The A Index is regarded as cotton’s
world price. Published by Cotlook Ltd,
the index is an average of the fi ve lowest
quotes for Middling 1 3/32’ upland
cotton from among 19 types traded
internationally. Prices represent offering
quotes, for delivery at Far East ports.
Uzbek Cotton Pricing

6
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
procurement system as insulating farmers from world price movements. This
characteristic is not necessarily intentional, but is defi nitely a consequence of
policy in Uzbekistan.
Figure 3
World, United States, and Uzbekistan cotton prices
Cents per pound
Sources: State procurement prices (SPP) from USDA’s attaché in Tashkent and the International
Cotton Advisory Committee (ICAC). U.S. cotton price from USDA, National Agricultural Statistics
Service, 2011. World cotton price from Cotlook Ltd. Exchange rates from the United Nations
Development Programme (2006) and TCX Investment Management Company B.V.
1999 2000 01 02 03 04 05 06 07 08 09
0
20
40
60
80
100
World
United States
Uzbekistan

7
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
More than 75 percent of Uzbekistan’s cotton is exported, so foreign
exchange rate policy is an important determinant of the NRA. Uzbekistan’s
Government controls capital fl ows and its exchange rate, and cotton and
gold are “centralized” exports. Centralization means that all foreign
exchange earnings from these commodities must be surrendered to the
government for conversion into local currency (the Soum) at the offi cial
exchange rate.8 At times, this policy mix has resulted in a signifi cant
implicit tax on the cotton sector as a result of the offi cial exchange rate’s
overvaluation. An International Monetary Fund (IMF) study of this policy
during its peak years (Rosenberg and De Zeeuw, 2001) found that exchange
rate overvalution transferred the equivalent of between 3.4 percent (1997)
and 6.7 percent (1999) of gross domestic product (GDP) from the cotton
sector to the government.
The goals of exchange rate policy can include expansion of trade, control-
ling infl ation, and assistance in achieving longrun development goals.
An undervalued exchange rate can promote exports, but an overvalued
exchange rate reduces incentives to produce tradable goods. It can even be
specifi cally employed as an export tax.9 However, an overvalued exchange
rate reduces import prices in domestic currency terms, reducing local infl a-
tionary pressure. Foreign exchange shortages are likely to result, which
can hinder imports but also provide a policy opportunity as sectors can be
targeted for assistance through preferential access to the limited supply of
foreign exchange.
Uzbekistan’s overvaluation fi rst became signifi cant late in 1996 and has fl uc-
tuated widely since then. Financial sector reforms starting in 2000 and two
large devaluations of the offi cial exchange rate in May 2000 and November
2001 helped bring the offi cial and parallel market exchange rates into align-
ment by 2003 (table 1). This culminated with Uzbekistan’s acceptance of the
currency convertibility obligations of Article VIII of the IMF’s Articles of
Agreement in October 2003.10 During 2008, however, the offi cial exchange
rate again began to diverge from the parallel market rate.11 Since then, over-
valuation ranged from 15 percent in 2008/09 to 48 percent in 2011/12.
When considering the interaction between the level of the SPP and the degree
of exchange rate overvaluation, note that cotton policy is probably not the
primary determinant of the offi cial exchange rate’s level or of its ratio to the
parallel rate. Instead, policymakers presumably set the SPP with knowledge
of the government’s goals with respect to the real exchange rate. While poli-
cymakers’ targeting of the real exchange rate may not be perfect, the rate has
followed relatively clear trends, so the export parity of the nominal SPP can
also be targeted to a large extent.
8Noncentralized exports have a 50
percent surrender requirement.
9Brazil’s multiple exchange rate sys-
tem during the 1930s is one example
(Barros, 2009). At that time, coffee
accounted for 70 percent of Brazil’s
exports, and coffee export earnings
were subject to repatriation at an
exchange rate higher than that applied
to many other commodity transactions.
10However, Pomfret (2009) and
others found that access to foreign
exchange in bank accounts is prob-
lematic, and even access to cash in the
form of domestic currency has been
diffi cult (Gemayel and Grigorian,
2005). In addition to the foreign
exchange price, the government has
administrative policies that can alter
the value of foreign exchange holdings,
examination of which is beyond the
scope of this study.
11The unoffi cial foreign exchange
market in Uzbekistan is referred to
here as the “parallel” market. Lindauer
(1989) assigned the term “black” mar-
kets to illegal parallel markets. Press
reports from Uzbekistan suggest that
parallel markets are currently illegal,
but may have been legal previously.
Exchange Rate Policy

8
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
Table 1
Uzbek cotton procurement prices and exchange rates
Marketing year
Procurement
(SPP)
Exchange rate
(offi cial)
Exchange rate
(parallel)
Parallel
exchange rate
market premium
Soums per ton —— Soums per U.S. dollar —— Percent
1999/2000 34,800 168 684 308
2000/01 52,000 331 903 173
2001/02 80,000 635 1,355 113
2002/03 126,000 918 1,153 26
2003/04 195,000 992 999 1
2004/05 225,000 1,069 1,067 0
2005/06 255,000 1,185 1,175 -1
2006/07 306,000 1,244 1,268 2
2007/08 352,000 1,293 1,344 4
2008/09 430,000 1,403 1,610 15
2009/10 454,850 1,537 2,101 37
2010/11 587,980 1,661 2,360 42
2011/12 496,020 1,850 2,730 48
SPP=State procurement price.
Sources: Procurement prices from USDA’s attaché in Tashkent and the International Cotton
Advisory Committee (ICAC). Exchange rates from the United Nations Development Programme
(2006) and TCX Investment Management Company B.V.

9
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
Accurate information about Uzbekistan’s economy is limited, and offi cial
statistics are often unreliable.12 Determining how economic policy affects
cotton production is also diffi cult to calculate due to the pervasiveness of
government intervention. The full policy impact on producers is a function
of numerous local administrative decisions and likely intentionally lacks
transparency. Uzbekistan, however, is not completely closed off from the rest
of the world. Procurement prices and exchange rates are available and can be
used to examine the relationship between Uzbekistan’s SPP and world prices
and to estimate whether procurement price policies result in an implicit tax
(or subsidy) on cotton production.
This approach is based on a comparison of the ratio of the SPP to world
prices against the comparable ratio—prices received relative to world
prices—for U.S. farmers. Uzbek farmers, like those in most low-income
countries, market their output as unginned seedcotton, so the SPP is a seed-
cotton price. In contrast, U.S. farmers gin their own cotton, market the fi ber
and seeds separately, and receive income from both products. U.S. and Uzbek
prices must, therefore, be adjusted to make them more directly comparable.
U.S. prices are adjusted by deducting the cost of ginning from the farm price
for fi ber and then adding in the value of the cottonseed sold separately. The
Uzbek SPP for seedcotton is adjusted to a fi ber-equivalent basis through divi-
sion by the proportion of fi ber typically derived from Uzbek seedcotton.13
U.S. and Uzbek cotton are of similar quality, so differences in export parity
may indicate whether pricing creates a tax or subsidy for Uzbek farmers.
Pomfret (2009) estimated Uzbek ginning costs for 2002, and Rudenko et
al. (2009) estimated the same costs for 2004. Both studies reported ginning
costs at around 10 cents per pound, comparable with U.S. costs for 2002
(USDA/ERS, 2012a). Estimated U.S. ginning costs for 1999-2010 are avail-
able from ERS’s cost and returns estimates; NASS’s farm price for cotton-
seed multiplied by the ratio of U.S. cottonseed to cotton fi ber production
provides an estimate of the additional value available to U.S. farmers from
their farm output (USDA/NASS, 2012; USDA/FAS, 2012).14 We assumed
that transportation costs to export markets were similar, and the distances to
Tianjin (China) from both Bandar Abbas (Iran) and Long Beach, CA, suggest
this assumption is plausible. Being landlocked—and entirely bounded by
neighbors that are themselves landlocked—Uzbekistan faces export impedi-
ments that do not affect U.S. exports, but evidence suggests that ocean freight
charges are lower for Uzbekistan, which is possibly offsetting.15
U.S. policy currently has little effect on the price U.S. cotton farmers receive
relative to the world.16 Therefore, the export parity of U.S. farm prices to
world prices should be representative of transportation costs and marketing
margins freely determined by market forces. Data from USDA’s Farm Service
Agency (FSA) support this conjecture. U.S. farm legislation mandates that
FSA survey cotton exporters annually to measure average marketing costs of
U.S. cotton to foreign markets. During 2004-09, these costs ranged between
19 and 24 percent of the Cotlook A Index (USDA/FSA, 2012). During the
same period, the export parity of U.S. farm prices averaged 74 percent,
12According to IHS Global Insight
(2011), “all economic data emanating
from the country [are] highly suspect.”
The IMF regularly notes weaknesses
and inconsistencies in Uzbek data that
hinder evaluation of economic perfor-
mance (IMF, 2007).
13Piggott and Wohlgenant (2002)
described the relationship between
the prices of processed goods jointly
produced from an input and the input’s
price. If the proportions of the joint
products are α and 1-α; the price of
seedcotton is A; the price of cotton seed
is B; the cost of ginning is C, and the
price of cotton fi ber is P, then: P = [A –
(1-α)B + C]/α. In Uzbekistan, average
ginning outturn is 32 percent (α). Here
we follow the widely accepted simplifi -
cation of dividing the seedcotton price
by the ginning outturn. This allows
direct comparison with the adjusted
U.S. price described in the text.
14Uzbek farmers are reportedly
permitted to purchase byproducts from
their ginned cotton at a discount, and
Velshwisch (2008) indicated that this
occurs at least in Khorezm, but other
authors suggest that the practice is
not widespread. The discount would
likely only add a few cents per pound
to the procurement price, changing the
calculations minimally. Rudenko (2008)
noted that Uzbekistan’s crushing indus-
try pays relatively little for the cotton-
seed it procures and is highly profi table.
15Verma (2008) found that transpor-
tation from Bandar Abbas to China
may be as much as 3 cents per pound
less expensive than from the West
Coast of the United States, or 4-5 per-
cent of the world price. USDA’s attaché
indicated that Uzbekistan’s supports
an excellent transportation infrastruc-
ture but, for our purposes, we simplify
calculations by assuming that higher
domestic costs offset Uzbekistan’s
external transportation advantage. The
need to transfer to a different gauge
rail system at the Commonwealth
of Independent States (CIS) -Iranian
border is an example of a higher cost
between farm and port.
16User Marketing Certifi cates (more
widely known as the “Step 2” program)
were determined by a World Trade
Organization (WTO) dispute panel to
act partly as an export subsidy. The
program was terminated after the
2005/ 06 marketing year. The Food and
Calculating the Export Parity
of Domestic Prices
Footnote 16 continued on page 10.

10
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
indicating an A Index premium of about 25 percent, consistent with FSA
cost estimates (fi g. 4). The reliability and speed of delivery of U.S. cotton
should allow U.S. cotton to command a premium on world markets, but this
premium is largely embodied in the farm prices of U.S. producers.
While adjusted U.S. farm prices for cotton averaged about 25 percent below
the world price during 1999/00-2010/11 (table 2), the Uzbek SPP averaged
about 50 percent below the world price when valued at offi cial exchange rates
and averaged about 60 percent below at parallel market rates, suggesting that
the offi cial procurement price substantially underestimates the value of Uzbek
cotton on world markets. Within this 12-year period, distinctly different sub-
periods can be discerned, with implicit taxes reaching their lowest levels
between 2005/06 and 2008/09.
Between 2005/06 and 2008/09, the SPP averaged only 44 percent below
the world price (at market exchange rates). Using the U.S. average discount
from export parity of 25 percent as the free-market cost of transportation and
marketing, the additional 19-percentage–point discount relative to the world
price received by Uzbek farmers indicated the implicit and explicit taxes
resulting from government-administered prices backed by border controls.
Our estimated taxes for 2000/01-2004/05 and the tax changes over time are
consistent with fi ndings from earlier studies,17 but new data indicate that the
lower taxes maintained during 2005-08 have recently increased.
17Rudenko et al. (2009) found that
Uzbek farmers received 66 percent of
the world price in 2004. Guadagni et
al. (2005) cited a fi gure of 77 percent
for 2004 and noted that implicit taxes
as a result of price controls fell with the
unifi cation of exchange rates. Pomfret
(2009) cited an estimate for marketing
year 2001 of 54 percent export parity,
and presented data consistent with a 37
percent parity using this methodology.
Agricultural Policy Research Institute
(FAPRI, 2005) estimated that the
program added a 2.9 percent premium
to the U.S. farm price relative to the
world price. This is similar to the
2.3-percentage-point relative decline
observed in the U.S. price between
2006/07 and 2009/10 compared with
between 1999/00 and 2005/06.
Footnote 16 continued f rom page 9
Table 2
Uzbek procurement and U.S. farm and world cotton prices
State procurement price (SPP)
Marketing year
Offi cial
exchange rate
Parallel
exchange rate U.S. farm price World price
U.S. cents per pound
1999/00 29.4 7.2 45.0 52.8
2000/01 22.3 8.2 49.8 57.2
2001/02 17.8 8.4 29.8 41.8
2002/03 19.5 15.5 44.5 55.7
2003/04 27.9 27.7 61.8 69.2
2004/05 29.8 29.9 41.6 53.5
2005/06 30.5 30.8 47.7 56.1
2006/07 34.9 34.2 46.5 59.1
2007/08 38.6 37.1 59.3 72.9
2008/09 43.4 37.9 47.8 61.0
2009/10 42.0 30.7 62.9 77.5
2010/11 50.2 35.3 81.5 165.0
2011/12 38.0 25.8 90.5 103.5
Note: Procurement prices for seed cotton converted to fi ber-equivalents by dividing by 32
percent ginning outturn ratio.
Sources: Procurement prices from USDA’s attaché in Tashkent and the International Cotton
Advisory Committee (ICAC). U.S. price from USDA, National Agricultural Statistics Service.
World price from the Cotlook A Index. Exchange rates from the United Nations Development
Programme (2006) and TCX Investment Management Company B.V.

11
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
Export parity differences could, under some circumstances, stem from
unavoidable differences in marketing costs—such as those introduced by
poor infrastructure investment (World Bank, 1999)—but this does not appear
to be the case with Uzbekistan. In developing countries, high costs can be
introduced by small-scale marketing systems with poor vertical coordina-
tion and integration. Underinvestment in storage facilities can lead to large or
potentially large post-harvest losses. Marketing in Uzbekistan happens on a
large scale, with few intermediate steps in the marketing chain and signifi cant
vertical coordination. Uzbekistan’s relatively arid climate and the seasonality
of its rainfall mean that precipitation during key points in processing and
shipping is comparable with the U.S. Southwest, where USDA has permitted
outdoor storage of cotton placed in the U.S. marketing loan program.
Investment in terminal facilities within Uzbekistan—and in railways
leading to export points—has been signifi cant as Uzbekistan has adjusted to
exporting outside the Commonwealth of Independent States (Cotlook, 2010).
There are ineffi ciencies embedded in Uzbekistan’s marketing system, but they
are largely a consequence of government-sanctioned or government-owned
monopolies and oligopolies at various stages. Uzbekistan’s rigid adherence
to export pricing based on widely published world prices provides a useful
monitoring mechanism to avoid FTC under-invoicing and off-shore profi t
capture. Greater returns, however, might be possible with more fl exible
marketing (Butler, 2005). Similarly, while a greater number of competitors
for transportation services for Uzbek exports might eventually lower costs,
government policy limits market entry. Such ineffi ciencies are ultimately
implicit taxes on the consumers of these services.
The Role of Marketing Costs
Figure 4
Export parity for U.S. and Uzbek cotton prices
Local farm price as percent of world
Sources: State procurement prices (SPP) from USDA’s attaché in Tashkent and the World
Bank. U.S. farm price for cotton and cottonseed from USDA’s National Agricultural Statistics
Service. U.S. ginning costs from USDA’s Economic Research Service cost and returns. U.S.
cotton and cottonseed production data from USDA’s Foreign Agricultural Service Production,
Supply and Distribution Online.
Marketing year
1999 2000 01 02 03 04 05 06 07 08 09 10 11
0
20
40
60
80
100
United States
Uzbek, market exchange rate
Uzbek, official exchange rate

12
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
The discount from export parity provides an estimate of total implicit and
explicit taxes—a negative NRA. This NRA can be disaggregated (table 3)
into impacts resulting from SPP policy (NRASPP) and exchange rate policy
(NRAE) (see Appendix for details). Both measures are negative in virtually
every year, suggesting that both the SPP and the exchange rate policy reduce
producer revenues. NRASPP is more stable in certain respects than NRAE
but fl uctuates consistent with the price-smoothing characteristic of the policy
noted earlier. For example, when world prices fell precipitously in 2008, the
NRASPP reached its lowest level of taxation, or 6 percent. Alternatively, when
world prices reached record highs in 2010, the NRASPP reached one of its
highest levels of taxation, or 52 percent. Taxation through NRASPP trended
downward during the early 2000s, stabilizing around 25 percent during
2004-07 before rising during 2010-12. NRAE remained at or close to zero
during 2003-07 but rebounded afterward, reaching 16 percent in 2011.
Nominal Rates of Assistance, by Policy
Table 3
Implied nominal rates of assistance (NRA), by policy (NRASPP and NRAE)
Marketing year
State procurement price
(NRASPP)
Exchange rate overvaluation
(NRAE)
Percent
1999/00 -26 -56
2000/01 -48 -33
2001/02 -43 -30
2002/03 -53 -10
2003/04 -46 0
2004/05 -26 0
2005/06 -28 1
2006/07 -21 -1
2007/08 -29 -3
2008/09 -5 -12
2009/10 -28 -19
2010/11 -52 -14
2011/12 -51 -16
Source: USDA, Economic Research Service calculations based on data from USDA’s attaché,
Cotlook Ltd, United Nations Development Programme (2006), and TCX Investment Management
Company B.V.

13
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
Subsidies comprise the other side of the balance sheet in an accounting of the
net impact of government policy on cotton, and input subsidies are report-
edly extensive in Uzbekistan. Abdullaev et al. (2009) found that subsidies
amounted to $400 million in 2004, or approximately 43 percent of the cotton
crop’s value, but asserted that they were lower in later years. Pomfret (2009)
lists subsidies to cotton producers during 2000-04 that were equivalent to
2.1-8.1 percent of GDP (or $251-$830 million).18 Guadagni et al. (2005) cited
extensive input subsidies ($290-$486 million) during 2000-04 for Uzbekistan
cotton farmers that extended to fertilizers, pesticides, and machinery services.
Rudenko et al. (2009) noted that farmers received credit at an annual interest
rate of 3 percent, compared with 16 percent market rates. Gilham et al. (1995)
noted that water was essentially free for farmers. Abdulleav et al. (2009)
noted that Water Users Associations (WAU) were established for fee-based
water distribution, but that farmers lacked the resources to pay for such
services, rendering the payment scheme ineffective.
Veldwisch and Spoor (2008) found that subsidized fertilizer was available to
farmers, with the amount distributed determined by their planted area and
expected yield. Farmers also were entitled to subsidized diesel, however,
the fuel was stored at Machine Tractor Parks (MTP) and distributed only
with MTP services. This study also noted that subsidized farm credit was
directed into bank accounts similar to settlement accounts used during the
Soviet Era. These same accounts were used to deposit payments for cotton
deliveries, while input costs from State-owned fi rms were deducted directly.
Input deliveries may have been partly diverted by offi cials for personal gain
and, according to numerous reports, it is very diffi cult if not impossible for
farmers to withdraw cash from these bank accounts. Rudenko et al. (2009)
found that the diffi culty accessing the accounts was perceived by farmers as a
constraint on cotton production.
Inconsistent subsidy estimates for Uzbek cotton partly refl ect the reluctance of
offi cials, traders, and farmers to share valuable information but also refl ect the
nature of Uzbekistan’s economy. Markets for credit and other inputs are poorly
developed, and, for example, bartering is commonplace. The conditions placed
on the use of subsidized inputs add diffi culty to calculating the actual value of
input subsidies. On the other hand, without complete markets, assigning a value
to State-provided services remains diffi cult and raises the possibility that their
actual value exceeds that assigned in the studies discussed here. When consid-
ering the high degree of taxation found in numerous studies, it is important to
fully appreciate the uncertainty of these calculations.
Cotton producers do see some benefi ts from this centralized system. First,
producers experience less inherent risk based on the price smoothing aspect
of the SPP determination. Uzbek farmers were protected from shifts in rela-
tive world prices during 2000-08 that resulted in a 36-percent decrease in
U.S. cotton planted area, compared with a 9-percent decrease in Uzbekistan.
Second, the system supports the maintenance and operation of the country’s
irrigation system. Managing one of the world’s more complex irrigation
systems is a not an insignifi cant task, and operation and maintenance issues
in Uzbekistan may be less critical than in some neighboring countries.
18Calculation based on GDP data
from IHS Global Insight.
Subsidies: Extensive and Hard To Measure

14
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
The data we analyzed only assessed the implicit taxation of cotton producers
through prices. The comprehensive level of net taxation or subsidization
depends on the impact of other policies as well. Earlier, comprehensive
studies indicated that shifts in net aggregate support for cotton were highly
correlated with pricing policy shifts. Shifts in net taxation also were corre-
lated with shifts in Uzbekistan’s cotton yields. Recent world price volatility
makes analysis of recent pricing policies more complicated but, combined
with recent cotton yield data, the shift in the relationship between domestic
and world cotton prices suggests that total net taxation increased after 2008.
Guadagni et al. (2005) and Pomfret (2009) attempted to assess the overall
taxation of Uzbek cotton. Guandagni et al. (2005) completed a detailed study
that included input subsidies, exchange rates, the SPP, taxes, and debt restruc-
turing. Debt restructuring was a signifi cant share of government support
in some years, which further complicated attempts to calculate support.
They determined that, on a net basis, taxation of the cotton sector reached 5
percent of GDP in 2000 and trended downward to 1.8 percent in 2004. As a
share of cotton farmers’ gross income, taxation trended from 50 to 20 percent
over the same period. Pomfret (2009) reported that net transfers from cotton
producers as a share of GDP fell from 12.3 percent in 2000 to between 2.1
and 5.4 percent during 2000-04.
Results were not entirely consistent across studies. Nevertheless, they show
that Uzbekistan’s net taxation of cotton declined in 2000-05. Note that the
estimated range of net taxation (20-50 percent) underestimates the total nega-
tive impact on cotton production in two ways. First, the costs to Uzbek cotton
farmers were not fully enumerated even in the studies detailed here. Research
by Rudenko (2008) and Velshwisch (2008) found that farmers also face
payment delays and skewed classing19 when marketing their crop. Second,
while subsidies partly offset taxes, they do so by introducing an additional set
of distortions. Therefore, resource misallocations can persist or can even be
amplifi ed. Abdullaev et al. (2009) provided concrete examples of this situa-
tion with respect to water management.
19“Classifi cation,” for our purposes,
describes the application of standard-
ized procedures for measuring the
physical attributes of cotton, analo-
gous to grading for grains. Gilham
(1995) noted that, for the early 1990s,
“Uzbekistan ginneries generally
produce a higher percentage of higher
grades than the percentage of these
grades that are procured,” and cited the
failure to pay farmers a premium for
higher quality as one of the reasons.
Taxation Offsets Subsidies for Uzbek Cotton

15
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
Since 2008, increased observed implicit net taxation through prices has
coincided with lower yields, similar in both respects to the years prior to
2004. While producers’ efforts and actual yields could vary with producers’
expected returns, an additional factor bolsters the link between taxation and
reported yields: implementation of the State order system during periods of
high taxation likely skews yield estimates for Uzbekistan.
Net taxation breaks the alignment of interests between cotton producers and
the State. While the State observes producers’ efforts to meet quotas, ulti-
mately the asymmetry of information favors producers. To some extent, the
asymmetry of information is less with respect to planting activity than with
respect to the effort expended to achieve optimal yields. This ensures that
some cotton area will be planted that producers later effectively neglect. Data
on planting also is likely to be infl ated, and the degree of infl ation will be
correlated with net taxation.
Pomfret (2000), USDA’s attaché to Uzbekistan (USDA/FAS, 2000), and
others have noted occasionally signifi cant smuggling of Uzbek cotton to
neighboring countries. Uzbek authorities cannot observe, for example,
every time farmers achieve above-average yields, creating opportunities
for producers to smuggle cotton out of the country. Thus, while we cannot
observe every aspect of net taxation, we can expect reported yields to vary
inversely with the level of taxation.
Uzbekistan’s post-independence cotton yields trended downward but
increased during the years when higher export parity indicated reduced taxa-
tion through pricing policy (fi g. 5). Yields jumped to a higher plateau during
2004-07, averaging 19 percent higher than the previous 4 years.20 20Yield variability in each period is
low, so the difference is signifi cant at
the 1-percent level.
Linking Policies To Cotton Production

16
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
Figure 5
Uzbek cotton yield and State procurement price (SPP) export parity
Kilograms per hectare
Note: Pre-1995 data is unreliable due to hyperinflation and the difficulties transitioning from
Russia’s ruble to a local currency.
Sources: Yield data from USDA’s Foreign Agricultural Service Production, Supply and
Distribution Online and ERS calculations (real SPP) based on data from USDA’s attaché in
Tashkent and the International Cotton Advisory Committee; the United Nations Development
Programme (2006); and TCX Investment Management Company B.V.
Marketing year
1987 89 91 93 95 97 99 2001 03 05 07 09 11 0
10
20
30
40
50
60
70
600
650
700
750
800
850
900 Export parity (at parallel
market exchange rates)
Yield
Percent

17
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
Other issues may affect cotton yield, but policy differences appear to have
accounted for the much of the difference between the two periods (1999-
2003 and 2004-07). Gilham et al. (1995) ascribed the post-independence
yield decline to the cumulative impact of limited crop rotation and reductions
in chemical and mechanical inputs driven by foreign exchange constraints.
The United Nations Environmental Program (UNEP, 2011) highlighted how
rising salinity and reduced streamfl ow have affected yields in Uzbekistan’s
downstream provinces. With Uzbek cotton completely irrigated, yield vari-
ability is low but not completely eliminated. Signifi cant problems with irriga-
tion supplies in 2000 and 2008 likely accounted for below-average yields. In
addition to reduced taxation, good weather in 2004 and 2005 helped sustain
yields. In 2010, growing conditions were very favorable, but yields remained
relatively low.21 Yields in 2010 were even lower than those in 2008, while
water intake data from the Amu Darya River indicate more than ample water
supplies for irrigation (Scientifi c Information Center, ICWC, 2012).
Post-2008 data show a rebound in estimated taxation but volatility in world
cotton markets must be considered when assessing this data. The price-
smoothing aspect of Uzbek policy means that plunging world prices in 2008
resulted in unprecedentedly low taxation levels through SPP (6 percent).
Shortly thereafter, record-high 2010 world prices drove the estimated NRASPP
to one of the highest taxation levels in this study (52 percent).22 Based on the
recent shift in cotton yields and a rebound in implicit taxation through prices
for almost every year since 2008, we concluded that net cotton taxation has
rebounded. Given that yields are currently at their pre-2004 level, net taxation
is likely also similar to its pre-2004 level, around 50 percent.
Data issues add a degree of uncertainty to our calculations. Subsidy values
are diffi cult to calculate, as are the costs of such administrative procedures
as delayed payments to farmers. Our research substitutes analysis of avail-
able aggregate information for a detailed accounting at each step along
Uzbekistan’s supply chain. But even such an accounting would hold signifi -
cant ambiguity, given the structure of Uzbekistan’s economy.
21On October 15, 2010, Cotton
Outlook’s Uzbekistan correspondent
stated that, “this season’s cotton pro-
duction…has benefi ted from excellent
growing conditions.”
22Note, however, that the export
parity of U.S. cotton shifted sig-
nifi cantly that year. Unprecedented
price volatility and levels character-
ized world cotton markets in 2010.
Farmers’ prices in both the United
States and Uzbekistan were largely
determined before the mid-season
peaks that drove the average 2010
world price to such an unprecedented
level. U.S. farmers missed these peaks
due to the timing of their market-
ing, and Uzbek farmers missed these
peaks due to the timing of SPP deter-
mination and government policy.
Other Infl uences on Yield

18
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
The variations in net taxation of cotton discussed here may be a result of policy
shifts in three broader aspects of Uzbekistan’s economy: food security, infl ation,
and the distribution of economic rents.
Negative rates of assistance for cotton production make grain production
relatively more attractive to producers and may play a role in food security
policy. Uzbekistan’s post-independence transition to hard-currency food
grain imports was an important factor in the reduction of cotton area during
the 1990s.23 The shift from managed food-grain shipments between repub-
lics within the Soviet Union to international market exchange increased
Uzbekistan’s risk in relying on grain imports. In the mid-1990s, when
Uzbekistan began taxing cotton production signifi cantly, its grain import
dependency ratio was 80 percent. By 1999/00, it had fallen to 12 percent,
where it remained through 2007/08 (USDA/FAS, 2012). The global trade
policy environment during this period also supported a reduced emphasis on
food self-suffi ciency as a result of low price volatility and a strong legacy of
increasingly liberal trade arrangements around the world.
In 2008, these trends were disrupted. Irrigation problems throughout
Central Asia sharply cut Uzbek grain production, and world grain prices
soared sharply. Uzbekistan’s main source of imported wheat—neighboring
Kazakhstan—briefl y imposed an export ban. In response, Uzbekistan
announced an area shift for the next season from cotton to grains. The
summer of 2008 also marked the reemergence of a parallel market premium
for U.S. dollars, reducing the cost of importing grain.
Infl ation and food security are closely linked in Uzbekistan, given food’s
large share of consumer expenditures. The large expenditure share for food
and lack of central bank independence makes controlling infl ation diffi -
cult with volatile food grain prices (Al-Eyd et al., 2012). Lacking effi cient
markets, there is greater reliance on administrative measures to limit price
increases, prolonging the time needed for stabilization. In 2008, Uzbek poli-
cymakers explicitly linked exchange rate policy to infl ation reduction (IMF,
2008) and, with world grain prices surging once again starting in 2010, this
situation likely continued.
Another likely concern for policymakers is the level and distribution of
economic rents. Acemoglu and Robinson (2012) characterize Uzbekistan as
an economy with extractive economic institutions, and highlight cotton in this
respect. IHS Global Insight observed that:
“Despite a high level of privatization, the country’s economy remains tightly
controlled by the top circle of leadership. Many, if not most, major private
entities are held by members, friends, and family of the ruling elite.”
Reduced cotton production has implications for government revenues through
excise taxes and to the benefi ciaries of government-protected monopolies
along the marketing chain. In the 1990s, when Uzbekistan’s net cotton taxa-
tion rose sharply, cotton accounted for 40-60 percent of export earnings. By
2001, this share had fallen to 25 percent and fell to 11 percent by 2010 as gold
23Uzbekistan’s cotton area fell 29
percent between 1987 and 1995. Cotton
area subsequently never exceeded its
1995 level by more than 1.5 percent but
trended downward 5-8 percent in 2000
and 2009.
Factors Driving Changes in Net Taxation

19
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
and natural gas grew in importance.24 Increased taxation of cotton produc-
tion, however, may sustain revenues provided by cotton, even as food security
concerns reduce the volume of cotton output and associated rents.
Uzbekistan is often characterized as pursuing import-substitution industri-
alization (ISI). Under ISI, natural resource-based export sectors are taxed
to fi nance investment in industry. Uzbekistan has had some success with
ISI, raising textile output, boosting investment in energy production, and
exporting cars. Policymakers may have decided to divert additional resources
to industrialization after 2008 but evidence for such a shift appears slim.
Looking forward, two factors likely will determine if Uzbekistan’s underlying
policy environment continues to support a cotton policy that includes higher
taxes on cotton producers:
• Water-related constraints on agricultural production, and
• Continued high world grain prices.
The balance between water availability and needs remains key to both
the intensity of Uzbekistan’s concern with food security and to local grain
production costs. Regional surface temperatures have been trending up,
while precipitation has been trending down in the catchment regions of
Uzbekistan’s irrigation system (UNEP, 2011). These environmental develop-
ments, combined with a growing reliance of upstream countries on winter
hydroelectric output, have placed greater pressure on summer streamfl ows
(Rakhmatullaev, 2010). As a result, Uzbekistan’s food security concerns will
remain in the forefront and local grain production costs may also face pres-
sure, suggesting that continued higher levels of cotton taxation could remain
to improve the relative attractiveness of grain production.
Trends in world food prices may ease infl ationary pressures over time.
USDA’s baseline forecasts include a reduction in real food grain prices
between 2010 and 2020 (USDA/ERS, 2012b). While relatively small, this
decline marks a shift from the large increases of 2005-11. Although prices
are not expected to return to the record lows seen before 2006, they will help
drive consumer prices down. Reduced infl ationary pressure could induce
Uzbekistan to reduce its exchange rate overvaluation and help moderate the
impact of food security concerns on cotton taxation.
24Other energy and oil products
accounted for 25 percent; and other
major products included food (10
percent), metals (7 percent), machinery
(6 percent), and chemicals (5 percent)
(State Committee of the Republic of
Uzbekistan on Statistics, 2011).

20
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
Uzbekistan is a landlocked, relatively low-income, and signifi cantly rural
nation that inherited a complex economic legacy from decades as a member
of the former Soviet Union. Its economic institutions emphasize centralized
control of the economy and the appropriation of rents from natural resources
for industrialization and for the benefi t of the governing elite. Relations
with neighboring countries—all former members of the Soviet Union—are
complex partly because of shared resources, such as the Amu Darya and Syr
Darya rivers. These circumstances are not expected to change in the imme-
diate future, and Uzbekistan’s cotton sector likely will grow slowly under a
relatively high tax burden.
Shifts in Uzbekistan’s cotton policy have exacerbated shifts in world cotton
markets since 2000. In the middle of the last decade—when world cotton
prices reached record lows relative to other commodities—Uzbekistan
reduced the economic burden on its cotton producers, resulting in higher
yields, output, and exports and driving world cotton prices even lower. Then,
as world cotton markets tightened and prices rose to new peaks, Uzbekistan
cut cotton output and exports, partly through increased taxation, adding
upward pressure on world prices.
USDA’s longrun baseline estimates for Central Asia indicate that cotton
production and exports likely will increase slowly over the next 10 years, and
the region’s share of world output is expected to decrease. The slow pace of
economic reform in Uzbekistan, which accounts for two-thirds of Central
Asia’s cotton output, will remain a key factor driving the modest outlook for
cotton in the region. The high level of net taxation illustrated here suggests
that signifi cant output gains could be achieved under alternative policies.
However, sustaining such gains may require further reforms and investments
resulting in more effi cient use of inputs, particularly water.
Conclusions

21
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
Abdullaev, I., C. De Fraiture, M. Giordano, M. Yakubov, and A. Rasulov.
“Agricultural Water Use and Trade in Uzbekistan: Situation and Potential
Impacts of Market Liberalization,” Water Resources Development 25(1):
pp. 47–63, 2009.
Acemoglu, Daron, and James Robinson. Why Nations Fail: The Origins of
Power, Prosperity, and Poverty, Random House, 2012.
Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela.
“Measuring distortions to agricultural incentives, revisited,” World Trade
Review 7: pp. 675-704, 2008.
Al-Eyd, Ali, David Amaglobeli, Bahrom Shukurov, and Mariusz Sumlinski.
Global Food Price Infl ation and Policy Responses in Central Asia,
Working Paper No. 12/86, International Monetary Fund, 2012.
Ayers, R.S., and D.W. Westcot. Water Quality for Agriculture, FAO Irrigation
and Drainage Paper 29 Rev. 1, Food and Agriculture Organization of the
United Nations, 1994.
Barros, Geraldo. “Brazil: The Challenges in Becoming an Agricultural
Superpower,” Brazil as an Economic Superpower? Understanding Brazil’s
Changing Role in the Global Economy, Lael Brainard and Leonardo
Martinez-Diaz (Eds.), Brookings Institution Press, 2009.
Butler, R. “Factors Infl uencing International Cotton Prices,” Cotton Outlook
Special Feature: Uzbekistan-2005 International Cotton and Textile
Conference, pp. 6-10, 2005.
Cotlook Ltd. “Uzbekistan,” Cotton Outlook 88(45): p. 7, 2011.
Cotlook Ltd. Uzbekistan: Weaving a Way Into the Developed World,
Merseyside, UK, 2010.
Djanibekov, Nodir, Inna Rudenko, John P.A. Lamers, and Ihtiyor Bobojonov.
“Pros and Cons of Cotton Production in Uzbekistan,” Food Policy For
Developing Countries: The Role Of Government In The Global Food
System, Case Study #7-9, 2010.
Environmental Justice Foundation. White Gold: The True Cost of Cotton,
London, UK, 2005.
Food and Agricultural Policy Research Institute (FAPRI). Impacts of
Commodity and Conservation Reserve Program Provisions in House and
Senate Reconciliation Bills, FAPRI-UMC Report #15-05, 2005.
Gemayel, E., and D. Grigorian. How Tight is Too Tight? A Look at Welfare
Implications of Distortionary Policies in Uzbekistan, Working Paper No.
05/239, International Monetary Fund, 2005.
References

22
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
Ghei, Nita, and Stephen B. Kamin. “The use of the parallel market rate as
a guide to setting the offi cial exchange rate,” Board of Governors of the
Federal Reserve System, 1996.
Gillham, F.E., T.M. Bell, T. Arin, G.A. Matthews, C. Le Rumeur, and A.B.
Hearn. Cotton Production Prospects for the Next Decade, Technical Paper
No. 287, The World Bank, 1995.
Guadagni, M., M. Raiser, A. Crole-Rees, and D. Khidirov. Cotton Taxation
in Uzbekistan: Opportunities for Reform, ECSSD Working Paper No. 41,
The World Bank, 2005.
IHS Global Insight. Country & Industry Forecasting, Uzbekistan, Economic
Growth: Recent Developments, 2011.
International Cotton Advisory Committee (ICAC). Uzbekistan cotton price
database, March 2011.
International Monetary Fund. Republic of Uzbekistan: 2006 Article IV
Consultation—Staff Report; Public Information Notice on the Executive
Board Discussion; and Statement by the Executive Director for the
Republic of Uzbekistan, IMF Country Report No. 07/133, 2007.
International Monetary Fund. Republic of Uzbekistan: 2008 Article IV
Consultation—Staff Report; Public Information Notice on the Executive
Board Discussion; and Statement by the Executive Director for the
Republic of Uzbekistan, IMF Country Report No. 08/235, 2008.
Isengildina, Olga, C.W. Herndon, and O.A. Cleveland. “Cotton industry
in Uzbekistan: Structure and current developments,” Beltwide Cotton
Conferences Vol. 1, pp. 297-301, 1998.
Josling, T., K. Anderson, A. Schmitz, and S. Tangermann. “Understanding
International Trade in Agricultural Products: One Hundred Years
of Contributions by Agricultural Economists,” American Journal of
Agricultural Economics 92(2): pp. 424-46, 2010.
Lindauer, David. “Parallel, fragmented, or black? Defi ning market structure
in developing economies,” World Development 17(12): pp. 1871-80, 1989.
Naumov, Y., I. Pugach, and Y. Yusupov. Uzbekistan’s Textile Industry: How
to Implement Development Potential? Policy Brief No. 1(14), United
Nations Development Programme, 2010.
Piggott, N.E., and M. Wohlgenant. “Price elasticities, joint products, and
international trade,” The Australian Journal of Agricultural and Resource
Economics 46(4): pp. 487-500, 2002.
Pomfret, Richard. “Agrarian Reform in Uzbekistan: Why has the Chinese
Model Failed to Deliver?” Economic Development and Cultural Change
48(2): pp. 269-84, 2000.
Pomfret, R. “Tajikistan, Turkmenistan, and Uzbekistan,” Distortions to
Agricultural Incentives in Europe’s Transition Economies, K. Anderson
and J. Swinnen (Eds.), The World Bank, pp. 297-338, 2008.

23
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
Rakhmatullaev, S., F. Huneau, P. Le Coustumer, M. Motelica-Heino, and M.
Bakiev. “Facts and Perspectives of Water Reservoirs in Central Asia: A
Special Focus on Uzbekistan,” Water Vol. 2: pp. 307-20, 2010.
Rosenberg, C., and M. De Zeeuw. “Welfare Effects of Uzbekistan’s Foreign
Exchange Regime,” IMF Staff Papers 48(1): pp. 160-78, 2001.
Rudenko, Inna. Value Chains for Rural and Regional Development: The Case
of Cotton, Wheat, Fruit and Vegetable Value Chains in the Lower Reaches
of the Amu Darya River, Uzbekistan, dissertation thesis, University of
Hannover, 2008.
Rudenko, I., J. Lamers, and U. Grote. “Can Uzbek Farmers get More for their
Cotton?” European Journal of Development Research 21: pp. 283–96, 2009.
Sadler, Marc. “Vertical Coordination in the Cotton Supply Chains in Central
Asia,” The Dynamics of Vertical Coordination in Agrifood Chains in
Eastern Europe and Central Asia, Working Paper No. 42, Johan Swinnen
(Ed.), The World Bank, pp. 73-114, 2006.
Scientifi c Information Center, Interstate Coordination Water Commission of
Central Asia. Database of the Amudarya river basin, 2012, http://www.
cawater-info.net/amudarya/.
Shtaltovna, A., A-K. Hornidge, and P.P. Mollinga. The reinvention of
Agricultural Service Organisations in Uzbekistan – a machine-tractor
park in the Khorezm-region, ZEF Working Paper Series 75, University of
Bonn, Center for Development Research, 2011.
State Committee of the Republic of Uzbekistan on statistics. Statistical
Review of Uzbekistan, 2011, http://www.stat.uz/en/reports/177/.
TCX Investment Management Company B.V. Personal communication, Jos
Kramer, Vice President, 2012.
United Nations Development Programme (UNDP). How to Boost Exports in
Uzbekistan: Policies and Institutions, Policy Brief No. 4, 2006.
United Nations Environmental Programme (UNEP). Environment and
Security in the Amu Darya Basin, report prepared on behalf of the
partner organizations of the Environment and Security Initiative, 2011.
U.S. Department of Agriculture, Economic Research Service (USDA/ERS).
Commodity Costs and Returns: Cotton 2010-11, 2012a, http://www.ers.
usda.gov/data-products/commodity-costs-and-returns.aspx.
U.S. Department of Agriculture, Economic Research Service. International
Baseline Data: Cotton Supply and Use, 2012b, http://www.ers.usda.gov/
data-products/international-baseline-data.aspx.
U.S. Department of Agriculture, Farm Service Agency (USDA/FSA).
Personal communication, Tom Bickerton, Economist, January 2012.
U.S. Department of Agriculture, Foreign Agricultural Service (USDA/FAS).
U.S. Team reports on Soviet cotton production and trade, FAS-M-277, 1977.

24
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
U.S. Department of Agriculture, Foreign Agricultural Service, Global
Agricultural Information Network. “Uzbekistan, Republic of: Cotton and
Products Update,” GAIN Report Number UZ0012, 2000, http://www.fas.
usda.gov/gainfi les/200010/50678605.pdf.
U.S. Department of Agriculture, Foreign Agricultural Service, Global
Agricultural Information Network. “Uzbekistan, Republic of: Cotton and
Products Annual,” GAIN Report Number UZ5004, 2005, http://www.fas.
usda.gov/gainfi les/200504/146119488.pdf.
U.S. Department of Agriculture, Foreign Agricultural Service, Global
Agricultural Information Network. Uzbekistan Cotton and Products
Updates, 1995-2011, 2011, http://gain.fas.usda.gov/Pages/Default.aspx.
U.S. Department of Agriculture, Foreign Agricultural Service. Production,
Supply and Distribution Online, 2012, http://www.fas.usda.gov/psdonline/.
U.S. Department of Agriculture, National Agricultural Statistics Service
(USDA/NASS). Quick Stats, 2011, http://www.nass.usda.gov/Data_and_
Statistics/Quick_Stats/index.asp.
Vel dw i s c h , G. Cotton, Rice & Water: The Transformation of Agrarian
Relations, Irrigation Technology and Water Distribution in Khorezm,
Uzbekistan, Inaugural-Dissertation zur Erlangung der Doktorwürde
der Philosophischen Fakultät der Rheinischen Friedrich-Wilhelms-
Universität zu Bonn, 2008,
Veldwisch, G.J.A., and M. Spoor. “Contesting rural resources: Emerging
‘forms’ of agrarian production in Uzbekistan,” Journal of Peasant Studies
35(3): pp. 424–51, 2008.
Verma, A. “Shipping Cotton to China: Cost and Infl uencing Factors,” Cotton:
Review of the World Situation 61(5): pp. 16-7, 2008.
World Bank. India Foodgrain Marketing Policies: Reforming to Meet Food
Security Needs, Volume II: Annexes and Statistical Tables, Report No. 18329-
IN, Rural Development Sector Unit, South Asia Region, August 17, 1999.

25
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
The discount from the export parity calculated here can be disaggregated into
separate impacts resulting from the State procurement price policy (NRASPP)
and the exchange rate policy (NRAE) using tools found in the agricultural
economics literature. Given that cotton exports are under a complete foreign
exchange surrender requirement and that control of domestic and interna-
tional trade means that domestic consumption of cotton fi ber is based on
world price (less a 15 percent discount, which essentially accounts for trans-
portation according to Naumov, et al.), the approach used by Anderson et al.
(2009) can be decomposed to illustrate the relative impact of the two policy
instruments. The divergence of the State procurement price (SPP) from the
world price can be analyzed as an export tax, and the divergence of offi cial
exchange rates (EO) from parallel rates (EP) can be treated as a separate addi-
tional tax. We defi ne *
W
P as the A Index in U.S. currency terms (PW) multi-
plied by 0.75 (i.e., 1-0.25, to adjust to a farm-equivalent), then:
*
** )1(
WP
WPxWP
SPP PE PEtPE
NRA ×−−×
=.
and,
OW
xEP
t×
−= *
SPP
1.
Following Anderson et al. (2009), when exchange rates are overvalued, the
price of exportables is reduced by the fraction ex. Since cotton is a “central-
ized” commodity with a 100 percent foreign exchange surrender requirement
for exporters,
P
O
xE
E
e=.1
Thus, the total border support nominal rate of assistance is,
*
** )1(
WP
WPxWPX PE
PEtPEe
NRA ×−−××
=, and,
SPPE NRANRANRA −= .
A complete accounting of assistance also would require data on subsidies.
While subsidies for inputs have been signifi cant in some accounting efforts,
input market distortions call these measures into question. The cost of late
payments and skewed cotton classing also factor into farmers’ costs, offset-
ting some of a given year’s input subsidies.
As noted previously, earlier studies with access to subsidy estimates from the
same period found that taxes on cotton exceeded subsidies, resulting in a net
taxation of cotton producers ranging from 20 to 50 percent. The appendix fi gure
1Note that the exchange rate in the
parallel market, EP, is not the long-
term equilibrium rate. The macro-
economic policies associated with the
overvaluation of the exchange rate
directly and indirectly drive the paral-
lel market rate below the long-term
rate, but it is the equilibrium rate in
the context of those policies (Ghei and
Kamin, 1999).
Appendix: Nominal Rate of Assistance (NRA)

26
Economic Policy and Cotton in Uzbekistan / CWS-12h-01
Economic Research Service/USDA
illustrates the relationships between world prices (PW), the effi cient farm equiva-
lent of world prices ( *
W
P), the SPP, and total taxes and net taxes.
The measure of total implicit and explicit taxes, or the NRA, is
*
*
W
W
PP
P
E
SPP −.
NRASPP can be translated into the terms used in the appendix fi gure and
simplifi ed to
*
*
W
W
OP
P
E
SPP −
.
Appendix figure
Price-based estimate of total and net taxation
Price (U.S. $)
Pw
P*w
SPP/Ep
Subsidies
Total of explicit
+ implicit taxes
Efficient marketing +
transportation costs
Net taxation