Identifying opportunities for value chain development in the Kenyan coffee sector

Abstract

By developing a model that describes the Kenyan coffee value chain, this study evaluates opportunities emanating from four scenarios representing productivity gains in the various value chain stages of the coffee sector and additional three scenarios reflecting shifts in market situations. Results show that productivity-enhancing policies have stronger effects on coffee output and export performance if they target the milling stage of the value chain. Export subsidy and favourable external marketing conditions also have stronger effects, distributed comparably across the various value chain stages. We, however, found that these gains in the coffee sector come at the expense of other cash crops such as cotton, tea, sugar and tobacco. The approach followed in this study is relevant as this trade-off between coffee and the other cash crop sectors may not be visibly shown using standard value chain approaches.

Keywords

cash crops value chain development sectoral modelling policy analysis scenarios

Introduction

Kenya produces a globally recognized top-quality coffee, 99% of which is exported to high-income countries such as Germany, Sweden, Belgium and the United States. However, the performance of the coffee sector has been declining continuously for the past couple of decades. The share of the sector in GDP declined from almost 10% in the 1970s to only 0.6% in 2005 (Chege, 2012), yet employing over 600,000 households (Ngeywo et al., 2014). Several factors contribute to this poor performance, among which (i) international price volatility, (ii) high direct and indirect taxation and (iii) low production and productivity are a few to name, leading to declines in exports. In addition, the low domestic value addition results in a very small share of the value of the end product going to farmers and actors at the lower stages of the value chain. Further, coffee farmers are elderly, with an average age of 55 years, who appear to be reluctant to adopt new technologies, leading to reduction in coffee production despite opportunities for profitability and access to expansion (Ngeywo et al., 2014).

The coffee value chain starts from the production of coffee cherries from the coffee trees (Figure 1). Access and availability of inputs (such as coffee seedlings, fertilizers and pesticides) for coffee cultivation is a critical component. The production of coffee cherries in Kenya is made by two major groups of actors: smallholder farmers and estates. There are about 700,000 coffee smallholder producers and 3217 estates in the country (RoK, 2016), cultivating over 128,000 and 42,000 hectares of land, respectively (Kuguru, 2016).

Figure 1.

Overview of the coffee value chain. Source: Author’s elaboration based on RoK (2016).

Once harvested, there follows primary processing which involves either dry processing or wet processing. In dry processing, the coffee cherries are directly dried by being laid out in the sun for a few days and then go through hulling machines. Alternatively, wet processing involves the pulling out of the cherries before going through washing and drying. On average, 91% of the coffee produced in the country is wet-processed (USAID, 2010).

The coffee beans are then milled. Both smallholder farmers’ cooperatives and estates supply the primary processed coffee to millers for secondary processing. Three main commercial millers operate in Kenya: the Kenya Planters Cooperative Union, Socfinaf and Thika Coffee Mills (Chege, 2012). The milling stage is generally the final stage of coffee processing taking place in Kenya before reaching the tinny domestic market or the principal export market. In very limited instances, the milled coffee is roasted for market supply, while most households have the tradition of roasting coffee at home for own consumption. Kenyan coffee is mainly sold on the export market, domestic sales representing less than 3% of total production (USDA, 2014), or about 1380 t (Chege, 2012).¹

Coffee operations by estates are vertically integrated; they grow, process and market coffee through the wholesale coffee auctions in Nairobi. Meanwhile, smallholder coffee producers market their coffee through cooperatives which undertake primary processing and contract secondary processing and marketing through the coffee auction on their behalf. The direct participation of smallholder farmers in the coffee value chain effectively does not go beyond bringing the coffee to the cooperatives.

Although its performance has generally been disappointing in recent periods, there is a possibility that the coffee sector can re-establish its status in the economy if the country can exploit its potential of producing over 200,000 t of coffee beans per year (Chege, 2012). Currently, the country is producing only 23% (or 46,000 t) of this potential, with disappointingly low average productivity of 0.27 t per hectare compared to 0.71 t in Ethiopia (Elias et al., 2017) and as high as 27.9 t in Saudi Arabia (Al-Abdulkader et al., 2017). Further, coffee is one of the main products the country exports in relatively raw form and the country would substantially gain if more processing and packaging of the products is achieved domestically.

In light of this background, this article examines and identifies opportunities for value chain development in the coffee sector of Kenya. Specifically, this study assesses the impacts of seven different interventions related to improving sectoral productivity and other external challenges (marketing margins, export prices and export policies) facing the coffee sector. This is done using an economy-wide general equilibrium modelling approach that incorporates several sub-sectors of the coffee value chain.

By providing analysis of value chain development in the Kenyan coffee sector under an economy-wide framework, this study contributes to knowledge in the following three ways. First, this is one of the very few works that integrate sector-focused value chain data with economy-wide data. Specifically, the study expands a 2014 social accounting matrix (SAM) for Kenya (Boulanger et al., 2016) by incorporating detailed data on the coffee value chain extracted from the study by Chege (2012). Second, unlike many other studies, the one at hand tests the relative effects of several policy and external changes to the coffee value chain on the overall performance of the sector. Third, while most other studies on the cash crop sector focuses on either productivity-enhancing policies or other public policies and external factors independently, this analysis brings these factors jointly and provides a comparative dimension to their effects. By doing so, it provides a number of insights in a bid to improve the contribution of coffee to farm incomes, export earnings and overall output.

The rest of the article is organized as follows: The succeeding section describes the modelling approach for the identification of interventions with stronger value chain development implications on the coffee sector of Kenya, including the data used, model adapted and design of value chain development scenarios. This is followed by a presentation of results and a discussion of findings, including a comparison of results with other related studies and a presentation of potential limitations of the modelling approach. The final section provides concluding remarks with some policy recommendations.

Modelling approach

Data for coffee value chain modelling

The 2014 SAM for Kenya

This study adapts an economy-wide general equilibrium model for Kenya. A more elaborated version of a highly disaggregated SAM representing the economic structure of the country in 2014 (Boulanger et al., 2016) is the main source of data for the coffee value chain analysis. The SAM includes 141 individual accounts grouped into 67 commodity types (including a single coffee commodity), a single trade and transport account, 44 production accounts (including a single coffee-producing activity), 14 separate factors of production (land, labour and capital), 6 household accounts, and a single enterprise, government, capital (savings–investments) and the rest of the world account each. SAM flows are valued at producer prices in the activity account and at market prices (including indirect commodity taxes and transactions costs) in the commodity account. Similarly, there is a distinction between home production for home consumption and marketed consumption, because own account consumption is valued at producer prices while marketed consumption is valued at market prices. Households are treated in the SAM as multi-product activities. The government account is linked to four tax collection accounts. For further on SAMs, see De Melo (1988).

Elaboration of the coffee value chain in the Kenyan SAM

The baseline SAM (Boulanger et al., 2016) did not have a detailed representation of the coffee value chain. We made an empirical contribution in this regard by further disaggregating the micro SAM such that the single ‘coffee’ activity in the baseline SAM is split into six coffee activities: (i) cultivation by estates, (ii) cultivation by smallholders, (iii) primary processing (dry processing and wet processing) by estates, (iv) primary processing by cooperatives, (v) coffee milling and (vi) coffee roasting. Based on the study by Chege (2012), the share of coffee production by smallholders is assumed 60% of total. Final product at one stage is intermediate input at the next corresponding stage. This intermediate input is combined with value-added (labour, capital and land factors) and other intermediate inputs to produce output. Information on other intermediate use (in aggregate) and value-added contribution at each stage of the coffee value chain is derived from the study by Chege (2012), as elaborated below.

Likewise, the single ‘coffee’ commodity is further disaggregated into four coffee commodities: (i) coffee on the farm gate, that is, coffee cherries, (ii) primary processed coffee by estates and cooperatives, (iii) milled coffee and (iv) roasted coffee. The 2014 SAM for Kenya shows that moist highlands and semi-arid households are the main producers of cash crops including coffee. However, there is no clear one-to-one factor–household relationship in the SAM.

In accounting for the distribution of value added in the coffee sector, the players in the coffee value chain in Kenya are characterized as smallholder farmers, estates, cooperatives and institutions in the coffee milling and roasting sectors. Farmers are basically only involved in the cultivation of coffee, while cooperatives are involved in primary processing of coffee produced by farmers. On the other hand, estates participate in vertically integrated coffee value chain activities involving cultivation, primary processing and secondary processing (milling).

Cost analysis at the coffee cultivation stage shows that farmers spend Ksh 35.0 worth of intermediate inputs per kilogram of coffee. With receipt of Ksh 76.5/kg of coffee from cooperatives, the value added at farmers’ level is estimated to be Ksh 41.5/kg. Using raw coffee worth Ksh 76.5/kg as intermediate input for primary processing, cooperatives receive Ksh 94.5/kg from millers, implying an average value addition at cooperatives level of Ksh 18.0/kg minus the cost they incur for primary processing.

Millers are critical players in the coffee value chain. Millers in Kenya incur average milling cost of Ksh 38.4/kg of coffee plus a payment of 94.5/kg for cooperatives. They receive a final market value of Ksh 604.96/kg of milled green coffee. The net value addition at the milling stage, therefore, is Ksh 465.06/kg, showing that the value added that accrues at the milling level is more than 10 times compared to that at the farmers’ level.² More remarkably, the price of roasted coffee (Ksh 3025/kg) is more than five times the price of green coffee (Ksh 604.96). However, intermediate input costs in the roasting process were not available from the study by Chege (2012); thus, informed judgement was applied to account for such costs in fully elaborating the SAM.

These differences in prices and contribution to production along the value chain are summarized in Figure 2. Values on the top left sections of each cell represent prices per kilogram of coffee, while those on the bottom right are percentage contribution of each value chain stage to overall sectoral value added, intermediate input use and total production. We see that the contribution of the coffee milling sector to total value added is by far the highest indicating the size of return going to coffee millers vis-à-vis actors at the lower stages of the value chain such as farmers and farmers’ cooperatives. This information is incorporated into the baseline SAM for Kenya and re-estimated for balancing using cross-entropy technique.³

Figure 2.

Prices and contribution to production along the value chain. Source: Own compilation based on Kenya’s coffee value chain studies.

The behavioural model

A computable general equilibrium (CGE) model elaborated in Aragie (2014) is used to analyse the conceptual model demonstrated in Figure 1 and to study opportunities for value addition in the coffee sector of Kenya. Such general equilibrium studies are based on clear causal mechanisms and transmission channels through which agriculture, including the coffee sector, interacts with the rest of the economy, thus allowing for a clear exposition and decomposition of results. The model applied incorporates some of the underlying features of the Kenyan economy such as the joint roles of semi-subsistence households in production and consumption since a considerable level of production of some agricultural commodities is consumed at home (Boulanger et al., 2016).

The coffee and other sectors are assumed to follow multilevel nested production structure, where smallholders and estates generally follow constant elasticity of substitution (CES) technology to combine primary factors and further determine the use of intermediate inputs. Consumers’ behaviour is defined by a two-stage consumption nesting structure such that households’ demand for commodities can reflect the source of commodities as defined in the SAM. At the bottom of the consumption nest is a CES demand system where notionally identical home-produced and marketed commodities are combined to provide aggregate consumption of the commodities. At the top of the nest, consumers maximize their utility from the consumption of a set of composite commodities (from the lower nest) subject to their budget constraints and a linear expenditure demand system derived from Stone–Geary utility function.

The trade block is central to the current analysis as coffee is principally an export commodity in the Kenyan economy. The optimal compositions of imported and domestically produced commodities in the domestic market are determined using CES functions following the Armington insight (Armington, 1969). The share of imports and domestic supply in composite supply of a commodity depends on relative import and domestic prices. Likewise, production is optimally allocated between the domestic and the export markets using constant elasticity of transformation functions based on relative prices in these two markets.

Value chain development scenarios

Seven scenarios were designed and modelled. The first four scenarios represent the introduction of innovations or new practices on each of the coffee value chain stages which would improve productivity, while the other three were trade policy and external shocks which would induce shifts in the market situation. Interventions to improve coffee productivity could be linked to past initiatives in Kenya under the Coffee Improvement Project II (SCIP II) and Stabilization of Export (STABEX)-funded projects. As noted by Karanja and Nyoro (2002), SCIP II was instrumental for smallholder farmers’ access to loans for productivity-enhancing inputs, advance payments, factory development as well as for training. STABEX funds were also used for the rehabilitation of coffee roads, electrification of coffee factories and in building of laboratories, all of which would improve performance in the coffee value chain.

The state of domestic market, trade policies and prices of coffee are instrumental for the performance of the sector. However, like many other developing economies, commodity markets in Kenya face high trade and transport margins (Renkow et al., 2006); the ad valorem tax equivalent of these costs reaching up to 28% of prices facing households. On top of this, smallholder farmers in Kenya are being charged a road tax of 1% of total sales value of coffee cherries to transport from their farms to their nearest cooperative factory (Karanja and Nyoro, 2002). Incentives and disincentives to the coffee sector also determine its competiveness and export performance. In the global coffee market, Kenya is a price-taker and is highly vulnerable to global price shocks (Lukanima and Swaray, 2014).

The choice of productivity changes in the first four scenarios is entirely arbitrary, but comparative rates are adapted to maintain some degree of comparability across the scenarios. The same assumption is followed by McDermott et al. (2005) in a study to identify opportunities for value addition in the New Zealand beef value chain. We assume a 10% reduction in trade and transport margin for scenario analysis following other studies on marketing margins in Africa (Teravaninthorn and Raballand, 2009; Van Campenhout et al. 2013). On the other hand, policy experiments on export subsidy to the sector and world prices of coffee are purely hypothetical and intend to provide indications on the distributional implications of these kinds of shocks across stages of the coffee value chain.

Scenario 1. A 10% productivity gain in the coffee cherry cultivation (at the farm) sector.

Scenario 2. A 10% productivity gain in the coffee primary processing sector.

Scenario 3. A 10% productivity gain in the coffee milling sector.

Scenario 4. A 10% productivity gain in the coffee roasting sector.

Scenario 5. A 10% reduction in trade and transport margin for the coffee sector.

Scenario 6. A 10% export subsidy for coffee milling and roasting sub-sectors.

Scenario 7. A 10% revival in world (export) prices of milled and roasted coffee.

These scenarios are run under short- and long-run assumptions characterized by differences on the extent to which the government account adjusts and the ease to which factors move across sectors. These will provide indications on how selected economic variables respond instantaneously or over a more flexible longer time period. However, these scenarios are not directly comparable with strict relative efficiency analysis and budgeting purposes.⁴ This shall be a follow-up exercise.

Results and discussion

Effects of productivity gains on the coffee sector of Kenya

Production effects

The potential implications of standardized (10%) changes in productivity resulting either from productivity-enhancing investment packages or from other positive exogenous changes (such as favourable weather or policy environment) on the various value chain stages of the coffee sector are studied and evaluated in terms of impacts on production, producer prices and export earnings. As discussed earlier, we examine the short- and long-run effects of these alternative investment options.

Model results show that productivity-enhancing policies at the milling stage of the value chain cause stronger gains in output due to (i) the relative size of this component of the coffee value chain and (ii) the strong backward and forward linkages it has with the rest of the value chain stages. We observe that a 10% increase in productivity in the coffee milling sector could lead to a 38.2% increase in the overall coffee sector in the short run, which increases by about fourfold over the longer run (Table 1). The milling sector alone expands by about 50% in the short run. This is not unrealistic as the sector is starting from a very small base even compared to the cash crop sector of Kenya. Nevertheless, these long-run gains in the coffee sector come at the expense of a 3.9% contraction in the other cash crop sector due to reallocation of factors of production towards the more efficient coffee sector as productivity improves. As a result, despite such marked expansion in the coffee sector of Kenya, the overall cash crop sector increases only by 3.1% and 8.9% over the short run and long run, respectively.

Table 1.

Percentage change in output.^a

	Baseline^b	10% productivity shock
		Scenario 1		Scenario 2		Scenario 3		Scenario 4
		Coffee cultivation		Primary processing		Coffee milling		Coffee roasting
		Short run	Long run	Short run	Long run	Short run	Long run	Short run	Long run
Change in composite output
Total agriculture	1990.93	0.06	0.09	0.02	0.03	0.52	1.68	0.03	0.03
Food agriculture	1652.80	0.00	0.01	0.00	0.00	−0.01	0.21	0.00	0.01
Cash crops	338.13	0.34	0.50	0.13	0.16	3.12	8.88	0.17	0.15
Coffee cultivation	2.66	7.01	12.50	0.68	2.11	27.50	162.26	0.18	0.68
Coffee primary processing	3.25	6.11	11.03	2.50	4.06	26.04	154.57	0.18	0.65
Coffee milling	18.55	4.27	8.15	1.79	2.94	50.00	172.62	0.23	0.62
Coffee roasting	4.69	0.67	1.31	0.28	0.48	6.63	17.78	12.04	12.27
Other cash crops	308.98	−0.02	−0.19	−0.01	−0.07	−0.18	−3.94	−0.01	−0.07
Change in value added
Total agriculture	1394.07	0.04	0.07	0.02	0.02	0.51	1.47	0.04	0.04
Food agriculture	1136.90	0.00	0.02	0.00	0.01	−0.04	0.32	0.00	0.01
Cash crops	257.16	0.23	0.29	0.10	0.11	2.95	6.56	0.20	0.17
Coffee cultivation	1.38	10.68	18.57	0.16	2.12	5.76	162.80	0.04	0.68
Coffee primary processing	0.55	2.16	5.55	12.66	14.33	37.44	150.04	0.23	0.62
Coffee milling	12.63	3.63	7.10	1.52	2.55	59.55	179.40	0.25	0.61
Coffee roasting	3.22	−0.01	0.05	0.00	0.01	0.11	1.83	15.28	15.48
Other cash crops	239.38	−0.01	−0.18	0.00	−0.06	−0.09	−3.73	0.00	−0.06

^aKenyan coffee value chain equilibrium model.

^bIn 10 million Kenyan Shillings (Ksh).

The cultivation stage of the coffee value chain offers the second-best avenue towards the development of the coffee sector. Study results demonstrate that productivity-enhancing investments on coffee cultivation could result in a 7.0% surge in output in the short run, increasing further by 5.5 percentage points over the longer run. This further propagates gains in production upwards along the value chain. We note that the production linkage declines gradually as one moves from the lower stage of the value chain to the upper stage. Primary processing and coffee roasting do not appear to be ideal areas of investment for coffee value chain development as the overall effects on the coffee sector are minute. This is principally due to very limited forward and backward linkages to the rest of the coffee sectors.

The lower section of Table 1 offers estimates of contribution to value added development. While changes in composite output reported above obscure the pattern of value creation since the uses of coffee as intermediate inputs in the succeeding stages of the value chain are included, changes in value added provide interesting patterns on returns to the various actors including farmers as providers of labour, land and capital.

Interestingly, changes in value added are markedly stronger in most cases than the composite output effects. However, the milling stage of the value chain remains the most effective entry point for the coffee sector development in the country, followed by support to productivity-enhancing technologies for farmers. The advantage with improving the milling sector is that it creates effective demand for coffee growers such that they will not be discouraged by low or volatile producer prices which have been characterizing features of the coffee sector in Kenya. Note also that the gain in value added is considerably stronger in the value chain stage experiencing the productivity gain and quickly declines for the production stages down or up this particular value chain stage.

Export, exchange rate and producer price index effects

Of the four stages of coffee output represented in our data, only two end up in the export market (milled coffee and roasted coffee). Kenya is not known for exporting roasted coffee; hence, there is no recognized level of export of roasted coffee in the export accounts of the country. We, however, impute a very small sector producing roasted coffee for the domestic and export markets for experiment purposes as we perceive that Kenya might emerge as a regional hub for trading a highly processed coffee.

Consistent with the results so far, improving the coffee farming and milling sectors is more rewarding in terms of the performance of coffee exports. Simulation results suggest that coffee exports could increase by more than 47.5% over the period of about 5 years, which could further surge by more than threefold over the longer period if coffee milling can become efficient by 10% (Table 2). Improving farmers’ productivity is also a feasible strategy as a huge part of the gain in production translates to the export sector which could grow by 4.1% and 7.8% over the short run and long run, respectively. Although this export performance of the coffee sector has no discouraging effect on overall agricultural exports of the country, the rest of the cash crop sector faces a contraction in exports mainly when coffee milling is supported due to a slight appreciation of the local currency. In general, the coffee sector policies examined so far are less likely to markedly affect the exchange rate in Kenya.

Table 2.

Percentage change in export, exchange rate and producer price index for coffee.^a

	Baseline	10% productivity shock
		Scenario 1		Scenario 2		Scenario 3		Scenario 4
		Coffee cultivation		Primary processing		Coffee milling		Coffee roasting
		Short run	Long run	Short run	Long run	Short run	Long run	Short run	Long run
Change in export and exchange rate
All agriculture ^b	357.99	0.20	0.27	0.09	0.10	2.36	5.47	0.13	0.11
Food agriculture	44.83	−0.03	0.01	−0.01	0.00	−0.39	0.17	−0.02	0.01
Cash crops	313.16	0.24	0.30	0.10	0.11	2.75	6.23	0.16	0.13
Coffee cultivation	–	–	–	–	–	–	–	–	–
Coffee primary processing	–	–	–	–	–	–	–	–	–
Coffee milling	16.57	4.59	8.76	1.93	3.16	53.44	183.34	0.18	0.60
Coffee roasting	2.62	1.05	2.05	0.44	0.75	10.24	26.98	18.52	18.87
Other cash crops	293.97	−0.02	−0.19	−0.01	−0.07	−0.18	−3.94	−0.01	−0.07
Exchange rate	1.00	−0.04	−0.03	−0.02	−0.01	−0.45	−0.65	−0.02	0.00
Change in producer price index
All agriculture	1.00	−0.17	−0.30	−0.03	−0.07	0.67	−0.91	−0.15	−0.17
Food agriculture	1.00	0.00	−0.03	0.00	−0.01	0.06	−0.59	0.01	0.00
Cash crops	1.00	−0.77	−1.22	−0.12	−0.25	2.77	−2.03	−0.73	−0.76
Coffee cultivation	1.00	−3.24	−5.05	0.49	0.00	18.67	−0.23	0.14	0.00
Coffee primary processing	1.00	−2.57	−4.14	−1.11	−1.60	16.59	0.08	0.13	0.01
Coffee milling	1.00	−0.42	−0.72	−0.18	−0.28	−3.23	−5.33	0.05	0.03
Coffee roasting	1.00	−0.51	−0.94	−0.22	−0.35	−4.51	−9.56	−6.81	−6.90
Other cash crops	1.00	−0.05	−0.03	−0.02	−0.01	−0.52	−0.65	−0.02	0.00

^aKenyan coffee value chain equilibrium model.

^bIn 10 million Kenyan Shillings (Ksh).

On the other hand, productivity-improving policies in the coffee sector have some noticeable effects on producer prices. The productivity policies are more likely to cause drop in producer prices of coffee, although there is evidence that these prices for coffee cherries and primary processed coffee could increase markedly in the short run if production in the milling value chain is targeted. In cases other than this, gains in production are associated with declines in producer prices (see Table 2).

Effects of trade policy and world price changes on the coffee sector of Kenya

Production effects

We further examine the potential roles of two more policy measures and one other exogenous shock on the performance of the coffee value chain. We specifically consider the implications of a 10% reduction in marketing margins and a 10% export subsidy to the coffee sector alone as two additional policy measures. In addition, we noted that the coffee sector of Kenya has been liable to risks of price failures in the international market. To examine the potential response of the sector to revival in the world price of coffee (or positive outlook in the international market), we simulate a 10% exogenous increase in export prices.

Compared to improvements in marketing margins, a 10% export subsidy has significantly stronger production effects on coffee, which could reach to about 26.3% in the short run at sectoral level (Table 3). This is consistent whether we consider composite output or output measured as value added. The very limited role of margins in value chain development is due to the relatively small size of the shock in relation to the coffee sector of Kenya. In comparison to the productivity shocks examined above, the marketing margin and export subsidy shocks have relatively balanced effects across the various stages of the value chain.

Table 3.

Percentage change in output.^a

	Baseline^b	10% positive shock on marketing margins, world price and export tax for coffee
		Scenario 5		Scenario 6		Scenario 7
		Margins for coffee		Export subsidy for coffee		World price for coffee
		Short run	Long run	Short run	Short run	Short run	Long run
Change in composite output
All agriculture	1990.93	0.01	0.02	0.36	0.70	0.69	2.90
Food agriculture	1652.80	0.00	0.00	0.00	0.06	−0.02	0.41
Cash crops	338.13	0.06	0.11	2.16	3.87	4.16	15.10
Coffee cultivation	2.66	0.75	2.14	25.15	73.71	47.11	316.55
Coffee primary processing	3.26	0.74	2.05	21.91	70.68	40.67	303.73
Coffee milling	18.55	0.73	1.73	19.94	74.14	36.97	321.33
Coffee roasting	4.69	0.90	1.08	22.48	70.62	41.72	303.11
Other cash crops	308.98	0.00	−0.04	32.26	69.54	61.63	291.46
Change in value added
All agriculture	1394.07	0.01	0.01	0.30	0.54	0.62	2.36
Food agriculture	1136.90	0.00	0.00	−0.02	0.10	−0.06	0.63
Cash crops	257.16	0.04	0.06	1.75	2.49	3.65	10.01
Coffee cultivation	1.38	0.19	2.15	7.27	73.80	12.47	317.64
Coffee primary processing	0.55	1.04	2.03	32.17	69.38	60.61	289.70
Coffee milling	12.63	0.73	1.61	2.77	74.45	4.69	324.99
Coffee roasting	3.22	0.27	0.29	31.84	69.38	59.87	289.70
Other cash crops	239.38	0.00	−0.04	36.04	69.14	69.57	287.22

^aKenyan coffee value chain equilibrium model.

^bIn 10 million Kenyan Shillings (Ksh).

Model results further show that world price of coffee is the most influential on the coffee sector. The effects are specifically stronger for coffee milling as it constitutes the lion’s share of coffee exports in Kenya. The strong backward and forward linkages and second-round interplays between the various stages of the coffee value chain specifically inflate the long-run effects of the increases on world prices for coffee. As a result, production by the rest of the cash crop sector could be compromised by as much as 7.7%.

Export, exchange rate and producer price index effects

While marketing margins remain weaker in affecting the export performance of coffee, the export subsidy policy option could result in significant response to coffee exports either in its green (32.3% or more) or roasted (7.9% or more) form (Table 4). Exports respond further strongly as the sector faces increases in international prices. However, these gains come at the expense of slight declines in exports of other cash crops, revealing potentially distortionary distributional implications.

Table 4.

Change in export, exchange rate and producer price index for coffee.^a

	Baseline	10% positive shock on marketing margins, world price and export tax for coffee
		Scenario 5		Scenario 6		Scenario 7
		Margins for coffee		Export subsidy for coffee		World price for coffee
		Short run	Long run	Short run	Long run	Short run	Long run
Change in export and exchange rate
All agriculture ^b	357.99	0.04	0.06	1.43	2.07	2.99	8.56
Food agriculture	44.83	−0.01	0.00	−0.25	0.05	−0.63	0.34
Cash crops	313.16	0.05	0.07	1.67	2.35	3.51	9.73
Coffee cultivation	–	–	–	–	–	–	–
Coffee primary processing	–	–	–	–	–	–	–
Coffee milling	16.57	0.71	1.78	32.26	69.53	65.67	308.45
Coffee roasting	2.62	1.31	1.59	7.88	14.04	36.04	74.40
Other cash crops	293.97	0.00	−0.04	−0.11	−1.54	−0.29	−7.68
Exchange rate	1.00	−0.01	0.00	−0.27	−0.24	−0.74	−1.25
Change in producer price index
All agriculture	1.00	0.02	−0.02	0.61	−0.58	1.40	−1.21
Food agriculture	1.00	0.00	0.00	0.06	−0.22	0.10	−1.13
Cash crops	1.00	0.06	−0.08	2.52	−1.82	5.88	−1.47
Coffee cultivation	1.00	0.55	0.00	16.06	−0.06	29.00	−0.44
Coffee primary processing	1.00	0.49	0.00	14.26	0.07	25.74	0.16
Coffee milling	1.00	0.02	−0.07	3.21	0.67	5.87	3.00
Coffee roasting	1.00	−0.51	−0.63	−9.26	−15.79	−3.39	−9.80
Other cash crops	1.00	−0.01	0.00	−0.31	−0.25	−0.85	−1.23

^aKenyan coffee value chain equilibrium model.

^bIn 10 million Kenyan Shillings (Ksh).

Export subsidy and increase in world price of coffee tend to result in short-run surges in producer prices of coffee along all value chain stages. These gains in producer prices are specifically bigger for coffee at farm gate and primary processed coffee by cooperatives and estates. However, these gains in producer prices are completely eroded over the long run whether world price shock or export subsidy is considered as the economy achieves greater adjustment.

How are results compared with related studies?

The coffee sector of Kenya is exposed to many challenges and constraints that limit its productivity: weather change, poor access to modern techniques of production, lack of access to productivity-increasing extension advice and poor access to inputs due to financial and credit constraints. These played their own share for the gradual decline in productivity. This article shows stronger effects of productivity-enhancing policies on the coffee sector output and export performance. Likewise, the positive and statistically significant roles of productivity-promoting interventions in coffee production are reputedly identified in the scientific literature on the case of Kenya and elsewhere (Minai et al., 2014; Nzeyimana et al., 2013; Were et al., 2002).

Further, the value chain analysis in this study implies a greater role for government trade policy for promoting coffee sector performance. This is in line with the significant and 1.4 elasticity of performance of the sector to a percentage improvement in the index of the coffee sector–related government policy identified by Kuguru (2016). Were et al. (2002) also show that the coffee sector in Kenya has been adversely affected by lack of export-promoting supports following the liberalization of markets under the regional economic blocs. This latter study further suggests that for maximum benefit from an export-led growth strategy, there is a need for incentives that help boost exports.

The study at hand also reveals a stronger effect of external marketing conditions on sectoral performance. This is in line with Orio (2015) and Were et al. (2002), who view world price volatility as a challenge to the coffee sector of Kenya. Likewise, Oiro (2015) identified an important role of foreign incomes in explaining Kenya’s coffee export performance. Hillocks (2001) also noted the prevailing price volatility in the world market as challenges facing the smallholder coffee sector in Africa, including Kenya, suggesting adapting the principles of integrated crop management to safeguard the livelihoods of coffee growers and contribute to the sustainability of the sector.

Model results further suggest a potential decline in the production of other cash crops that would come because of competition for resources and a transition of factors of production such as labour and land towards the coffee sector experiencing productivity gains. This finding is consistent with the literature on the potential discouraging implications for policies on food crops that promote cash crop production (Anderman et al., 2014; Von Braun, 1995). Specially, Von Braun (1995) investigated the case of Kenya and identified a negative and statistically significant relationship between production of cash crops and production of food crops due to competition for resources. This same rationale holds for the apparent decline in the performance of other cash crops as the productivity of the coffee sector of Kenya is simulated to increase.

Model results indicate weaker effects of productivity-increasing investments at the coffee roasting stage on actors down the supply chain given the current structure of the coffee sector in Kenya, since returns are flowing to value-added inputs outside farming due to the intensive use of such inputs at this stage. This finding does not fully disregard the relevance of domestic value addition by increasing the share of roasted coffee. Beyond improving productivity at the coffee roasting stage, instituting incentives that would (i) narrow the high margins between farm gate and distribution prices of coffee and (ii) increase the share of coffee roasted domestically should rather be the priority. High cost of coffee roasting infrastructure, lack of technical expertise, high tariffs on processed coffee and other entry barriers in importing countries have been identified by many (Chege, 2012; Ponte, 2002) as challenges to coffee roasting locally, and relaxing such constraints would promote local processing and allow coffee growers benefit more.

Potential limitations of the study

Although this study provides a number of insights on the role of policies and external factors in coffee value chain development in Kenya, there may be a number of limitations in light of which one should interpret the results. The study has not accounted for challenges and constraints facing the coffee sector of Kenya, which would alter sectoral performance. Studies indicate that the genetic resources of coffee are increasingly under threat due to the depletion of forest ecosystems (Gole et al., 2002). The climate is predicted to continue changing in the coming decades, with a huge impact on coffee production. Noticeable effects of climate change have already been documented in coffee-producing regions such as Kenya (Adhikari et al., 2015; Krishnan, 2017). There are also evidences of discouraging policy and regulatory contexts for coffee sector actors in Kenya, which this study did not fully account for. Kuguru (2016) shows that a 0.7-unit increase in the score of the government’s policies will have a unit increase in coffee industry performance in the country, implying the need for reforming its policies and regulations to rejuvenate coffee industry performance.

Further, despite the relevance of the approach adapted for this study, mainly in terms of studying the coffee sector within the overall economic framework of the country rather than in isolation, the method is prone to challenges when it comes to the number of behavioural parameters (such as elasticities) which are not readily available in most cases and analysts have to make informed judgement or take those from other empirical literature. However, most, including Hertel et al. (2007), maintain that CGE models remain the preferred tool for ex ante analysis of policies with effects on the wider economy.

Although the potential limitations discussed above imply future areas of research, it is safe to say that this study already contributes to knowledge in the area of value chain analysis of strategic crops in several ways. First, it contributes to knowledge by integrating sector-focused value chain data with economy-wide SAM. Second, the economy-wide model adapted here and augmented to capture value chain analysis allows stakeholders in the Kenyan coffee industry and beyond to understand the impacts that changes in farm practice, markets structures (domestic and international) and tax policies are likely to have on the industry itself and at a national level. Third, unlike most other studies which analyse selected sectors in isolation from the rest of the economy, this study examines the likely sectoral and economy-wide effects of changes. As such, it provides several insights in a bid to improve the contribution of coffee to farm incomes, export earnings and overall output.

Conclusion and policy recommendation

We have developed a model that describes the Kenyan coffee value chain. Four scenarios representing productivity gains in the various stages of the coffee sector and additional three scenarios reflecting shifts in market situations are considered. Results from the value chain model are presented to encourage debate about the future shape of the Kenyan coffee sector and guide directions for sectoral investments in helping to ensure these changes occur. The study shows that productivity-enhancing policies have stronger effects on the coffee sector output and export performance if they target the milling stage of the value chain, followed by the cultivation stage. Productivity-enhancing investments in the primary processing and coffee roasting stages do not appear to be ideal priority areas for value chain development as the overall effects on the coffee sector are minimal. This is owing to poor forward and backward linkages the coffee roasting stage in particular has under the current structure of the sector. This rather implies the role for incentive packages that would (i) narrow the high margin between farm gate and distribution prices of coffee and (ii) increase the share of coffee roasted domestically. Export subsidy and favourable external marketing conditions also have stronger effects, distributed comparably across the various value chain stages.

We, however, found that these gains in both production and export of coffee come at the expense of some declines in the performance of other cash crop sectors. The approach followed in this study is relevant as this trade-off between coffee production and the performance of other cash crop sectors may not be visibly shown using standard value chain approaches. While this trade-off in production reveals potentially distortionary distributional implications for the coffee policies examined, there is no evidence of net losses in the performance of the overall cash crop or the agricultural sector.

The results of the study imply a need to unlock challenges and constraints facing stakeholders along the coffee value chain. The government should enact a diverse range of policies that would reduce undesirable controls on production and marketing and provide an enabling environment for enhanced participation of the private sector. Although the government of Kenya has progressed a lot in this regard, it has to further liberalize the sector by separating the role of coffee marketing from its regulation. Necessary support should be provided to the financial sector to guarantee debt relief on non-performing loans owed by coffee farmers and further promote access to affordable credit to coffee growers.

Footnotes

Acknowledgements

This research was conducted in the context of Monitoring and Analysing Food and Agricultural Policies (MAFAP; ) Program implemented by the Food and Agricultural Organization (FAO), in collaboration with the Organisation for Economic Co-operation and Development (OECD). The study benefits from inputs provided by Caroline Demanet, FAO, Rome.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study is financially supported by the Bill and Melinda Gates Foundation, United States Agency for International Development (USAID), the Netherlands and Germany.

Notes

References

Adhikari

Nejadhashemi

Woznicki

(2015) Climate change and Eastern Africa: a review of impact on major crops. Food and Energy Security 4(2):110–132.

Al-Abdulkader

Al-Namazi

AlTurki

, et al. (2017) Optimizing coffee cultivation and its impact on economic growth and export earnings of the producing countries: the case of Saudi Arabia. Saudi Journal of Biological Sciences in press.

Anderman

Remans

Wood

, et al. (2014) Synergies and tradeoffs between cash crop production and food security: a case study in rural Ghana. Food Security 6(4):541–554.

Aragie

(2014) Climate change and semi-subsistence agricultural households. Thesis submitted in partial fulfilment of the requirements of the degree of Doctor of Philosophy, Oxford Brookes University, UK.

Armington

(1969) A theory of demand for products distinguished by place of production. Staff Papers-International Monetary Fund 16(1):159–178.

Boulanger

Dudu

Ferrari

, et al. (2016) Are input policies effective to enhance food security in Kenya? A CGE analysis. Paper prepared for the 5th international conference of the African Association of Agricultural Economists, 23–26 September 2016, Addis Ababa, Ethiopia.

Chege

(2012) Value Addition in Coffee Industry in Kenya: Lessons from Cut Flower Sector. Kenya Institute for Public Policy Research and Analysis (KIPPRA) Nairobi, Kenya, ICBE-RF Research Report No. 21/12.

De Melo

(1988) SAM-based models: an introduction. Journal of Policy Modeling 10(3):321–325.

Elias

Worku

Mathewos

(2017) Technical efficiency of smallholder coffee farmers in Gedeo Zone, Southern Ethiopia: A stochastic frontier approach. Agricultural Science Research Journal 7(4):147–153.

10.

Gole

Denich

Teketay

, et al. (2002) Human impacts on Coffea arabica gene pool in Ethiopia and the need for its in-situ conservation. In: Engels

JMM

Rao

Brown

AHD,

, et al. (eds) Managing Plant Genetic Diversity. New York: CABI, pp. 237–247.

11.

Hertel

Keeney

Ivanic

, et al. (2007) Distributional effects of WTO agricultural reforms in rich and poor countries. Economic Policy 22(50):289–337.

12.

Hillocks

(2001) Coffee: is it still a viable cash crop for smallholders in Africa? Outlook on Agriculture 30(3):205–211.

13.

Karanja

Nyoro

(2002) Coffee Prices and Regulation and Their Impact on Livelihoods of Rural Community in Kenya. Nairobi: Tegemeo Institute of Agricultural Policy and Development, Egerton University.

14.

Kuguru

(2016) Effects of government policies on performance of coffee industry in Kenya. IOSR Journal of Business and Management 18(6):53–60.

15.

Lukanima

Swaray

(2014) Market reforms and commodity price volatility: the case of East African coffee market. The World Economy 2014:1152–1185.

16.

Renkow

Hallstorm

Karanja

(2006) Rural infrastructure, transaction costs and market surpluses in Kenya. Journal of Development Economics 73(1):349–367.

17.

RoK (2016) Report of the national task force on coffee sub-sector reforms. Presented to H.E. Uhuru Kenyatta, C.G.H. President of the Republic of Kenya (RoK) and Commander-in-Chief of the Kenya Defence Forces. Available at: http://www.agricoop.info.ke/files/downloads/REPORT%20OF%20THE%20NATIONAL%20TASK%20FORCE%20%20ON%20COFFEE%20SUB-SECTOR%20%20REFORMS%20FINAL.pdf (accessed January, 2017).

18.

Krishnan

(2017-06-28) Sustainable coffee production. Oxford Research Encyclopedia of Environmental Science. Available at: http://environmentalscience.oxfordre.com/view/10.1093/acrefore/9780199389414.001.0001/acrefore-9780199389414-e-224 (accessed 9 February 2018).

19.

McDermott

Dooley

Smeaton

(2005) Identifying opportunities to add value to the New Zealand beef industry: a modelling approach. Proceedings of the New Zealand Society of Animal Production 65:252–255

20.

Minai

Nyairo

Mbataru

(2014) Analysis of socio-economic factors affecting the coffee yields of smallholder farmers in Kirinyaga County, Kenya. Journal of Agricultural and Crop Research 2(12):228–235.

21.

Ngeywo

Shitandi

Basweti

, et al. (2014) Effect of farm succession on coffee production in Kisii county Kenya. International Journal of Agricultural Extension, 2(3):183–191.

22.

Nzeyimana

Hartemink

(2013) Coffee farming and soil management in Rwanda. Outlook on Agriculture 42(1):47–52.

23.

Oiro

(2015) Real exchange rate volatility and exports in Kenya: 2005–2012. Journal of World Economic Research 4(5):115–131.

24.

Ponte

(2002) Brewing a bitter cap? Deregulation, quality and the reorganization of coffee markets in East Africa. Journal of Agrarian Change 2(2):248–272.

25.

Robinson

Cattaneo

El-Said

(2001) Updating and estimating a social accounting matrix using cross entropy methods. Economic Systems Research 48(1):47–64.

26.

Teravaninthorn

Raballand

(2009) Transport Prices and Costs in Africa: A Review of the Main International Corridors. Washington, DC: World Bank.

27.

USAID (2010) Kenya coffee industry value chain analysis. Profiling the actors, their interactions, costs, constraints and opportunities, United States Agency for International Development. East Africa Trade and Investment Hub - Nairobi, Kenya.

28.

USDA (2014) Kenya Coffee Report. GAIN Report, United States Development Agency (USDA), Washington, D.C.

29.

Van Campenhout

Pauw

Minot

(2013) The Impact of Food Price Shocks in Uganda: First-Order versus Long-Run Effects, IFPRI Discussion Paper 01284, Washington, DC.

30.

Von Braun

(1995) Agricultural commercialization: impacts on income and nutrition and implications for policy. Food Policy 20(3):187–202.

31.

Were

Ndung’u

Geda

, et al. (2002) Analysis of Kenya’s Export Performance: An Empirical Evaluation. KIPPRA Discussion Paper No. 22, Kenya Institute for Public Policy Research and Analysis, Nairobi, Kenya.