As a result, value chain analysis is undertaken to varying degrees of rigour and costs. It is no use pretending the results would be the same. We like to differentiate between snapshots and chain analysis. Snapshots are a very useful beginning. Of course, it is possible to overdo analysis because that makes it rather academic. Some spend $150,000 on a thorough value chain analysis while others budget $60,000 for 6 chains. Both are called value chain analysis but all other things being equal, there will be very significant differences between the two, However, higher cost is no guaranttee that the value chain aanlysis will be good.We have just seen some that cost $300,000 in the pacific on the AAACP/EU project that are frankly horrific. There is the inescapable conclusion that many people do not know what is possible with VCA,
Some people, even those representing international development agencies, think that a pictorial depiction of a supply chain amounts to value chain analysis. They even use words like 'mapping the value chain' as if that amounted to something significant.
It is actually not all that important to gather data and represent it as a value chain without analysis. Gathering data is not much use unless it is analysed and then it becomes only as good or astute as the analysis. There is a great deal that can be read by an expert from statistics that is simply not possible for non experts. Too many undertaking value chains think that they are a matter of routine that can be undertaken by local teams without expert guidance.
Also merely presenting costs incurred by major actors along the value chain is not enough since it is only when you view that information in the context of the whole chain that you can identify weaknesses and strengths. Simply accumulating hundreds of pages of report on stages of the chain, as one recent report in south-east Asia recently did, actually makes it impossible to use the information. It can often end up with hundreds of pages on value chains without a single complete chain with analysis.
As an example, we have seen a number of attempts at value chain analysis on the main commodities in Cambodia (more than seven that we have read), some were better than others but we are not completely happy about any of them. They all appear to give figures that show that farmers are making the healthiest profits per ton along the chain and show very narrow trader and retail margins (too narrow even to allow for normal wastage rates). This is not unique to Cambodia and we suspect the answer may lie in the difference between nominal and actual prices. Researchers are often given nominal figures and do not know enough to translate them into actual traded figures. The latter requires qualitative interviewing that is searching. A farm gate price may thus be nominally based on particular specifications and any departures, such as moisture content, and quality have to be deucted. A $260 per ton nominal farm gate price can easily end up as $190 per ton without taking account whose weights and measures are used.
Paradoxically, a number of studies give costs and margins at each major stage in the value chain, as in the cases above, but fail to see the logic of what they are claiming aside from the fact that they are often impossible. Analysts are sattisfied in observing that the chains are low value ones, viable only because of the volumes involved, and propose to make them better quality with more added value without paying sufficient heed to the fact that the chains service. They are difficult to tinker with because just making them more expensive with grading and certification is often not what the existing consumers want and higher purchasing power is usually growing slowly if at all. Value chains reflect circumstances, they cannot determine or fashion the latter, studying them is a useful way of understanding the circumstances and dynamics within which they are operating
While value chain analysis is a valuable tool, it is important to ensure a pro poor focus by including analysis of farmer livelihood systems. The term 'portfolio analysis' is the one we prefer. It entails looking at the portfolio of farmer allocations of resources to identify outputs and the contributions they make. By including farmer analysis within the overall value chain approach, intervention can aim to increase farmer livelihood standards. There is nothing inherent in value chain anlysis that is pro poor. It can just as easily be anti poor. If it is to be pro poor, there needs to be a focus on small farmers in the study and the strategy.
The entire process also needs to be based on a participatory approach of the sort adopted for the EU-ACP programme in which, both farmer analysis and value chain analysis are undertaken through workshops that leave ownership of the resulting strategy with framers. We have also since tried to get it applied in Cambodia on the AusAid CAASP design.
Mistakes in value chain analysis such as through using the wrong farm gate figures can lead to a false assumption that farmers do not need additional finance and instead transfer financial assistance to traders and millers. This change in focus instead of being poverty neutral can actually be anti-poor. We have read at least three reports on a South East Asian country for leading donor agencies that have made this error and have explicitly wrongly stated that farmers do not have a financial issue to be dealt with, a patently ludicrous conclusion.
We offer to undertake value chain analysis to specified levels depending on budgets (some field work is inevitable). It is better to know how costly different levels are. However, the chain is not worth much as a nominal one unless it states that it is that and the assumptions that underlie nominal status.
Example: Agriculture Sector Cambodia
When looking at any country, it is a good idea to start with a snapshot of the sector as a whole. Despite a lot of Agencies spending a lot of money over time even if it has been in modest installments, none undertook an overview of the sector in Cambodia. Instead they looked at particular chains while missing or rather avoiding the issue of porous borders and poor statistics.
If you have a good value chain snapshot to start with, there is so much information you can glean from it. Our current estimate of the Value Chain for Agriculture in Cambodia is:
$2.3 billion produce at farm gate level, with $1.1 billion going across porous borders to be processed with a FOB value of $1.9 billion while $1.5 billion is the domestic processed value with direct exports of $0.2 billion and the processed value ending up as retail value of $1.62 billion. However, there are a lot of assumptions and estimates. It is suprising that none of the many aid projects in the country have ever published estimates. Also disappointing is the level of coverage in nearly all countries on farm gate price data.
Some excellent value chains have been undertaken but not so many that you dont have to search for them. UNDP in Cambodia has a good cassava study undertaken recently and the World Bank commissioned an excellent one on rice in Madagascar some years ago. We have prepared value chain analysis for rubber in Indonesia, Malaysia and Cambodia and for horticulture in Sri Lanka.
It is useful to have benchmarks as reference points. In Europe, a ratio of 3:1 between retail and farm gate is considered efficient. A chain with 5:1 usually leads to farmers mounting protests against rampant exploitation. If a report claims a soft commodity value chain in a developing country to be 2:1, some scepticism is in order. If all chains are reported at such a ratio, the scepticism is all the stronger.
We undertake value chain analysis all the time for our own analysis if not for clients. We have been engaged with coir since 1976. But it has taken a long time to obtain a clear understanding of a product chain from farm gate to retail. In the way of illustration, the export chain and price make up for coir geotextiles in the form of stitched blankets:
A soft commodity that is to be processed into a product for industrialised countries can often have a 20 fold mark up from field level to retail price. It is also not unusual to see farm gate prices at 25% of FOB value. Farmers who process copra, in the way of an illustration, get around 25% of FOB value.Our experience indicates this ratio to be common in such cases.
Example: Vietnam Rice
An old study, itis not easy to use a current one without offending people but although the figures are one-third of what they are today, the price mark ups and margins indicated are quite typical of current studies on rice.
|Producer || ||Assembler ||Miller|| |
| || || || || || |
| || |
|US$ per ton||Retailer|
| || |
In the example above, producers sell at 87.2, assemblers at 93. millers at 153, wholesalers at 165 abd retailers at 172. Milelrs losses during operation are included. The chain ends up at under a ratio of 2 between farm gate and retail. But there is no allocation for other wastage and despite the fact that the chain aside from the farmer is handling many tons, unless it was a huge output, it is difficult to see how the chain would be viable.
It amazes us that donors and state authorities accept such findings without realising that they raise more questions that they pretend to answer.
Example: Cambodia Rubber
A number of attempts have been made to undertake a value chain for rubber in Cambodia. The latest one was commissioned by IFC in 2008. We did one in 2005 for AFD but when published, all the reports, including the AFD study we worked on appear to have shied clear of actually laying down the chain. Instead, they covered various aspects in detail and sometimes almost all aspects, but no chain was outlined in a clear and simple way, so here it is:
|FOB Singapore||1,200.0 |
|Net FOB Cambodia||1,052.24 |
|Factory price||1,035.39 |
|Factory gate||775.39 |
|less informal||116.31 |
|Factory pays||659.08 |
|DRC 60%||395.45 |
We trust the above because it is our calculation. All figures were cehecked in the field and all had more than two independent confirmations. That makes the value chain study good and reliable. We have the data to back that up with detailed data on every stage of the chain. what is more, we have comparable data on Indonesia and Malaysia.
Example Kyrgyz potatoes
Production of potatoes in Kyrgystan caters for neighbouring markets, mostly Khazakstan. It is very decentralised and there is little official support although a few NGOs are trying to help. The old days of public sector planning are represented by large disused warehouses in various states of disrepair, The result is a rudimentary system catering to the needs of neighbouring countries but with few profits.
Farm gate price of 4.50 soms per kg at Karakol, a local market price of 5.50 and a Bishkek market price of 7.00. Leads to Bishkek market selling price of 10-12 and a Kazhakstan market price of exported potatoes pf 16-17.
Example Solomon Islands Copra to Manila
The value chain for copra produced on small islands, whether they are in the Pacific or the Caribbean, is determined by similar factors. The selling price is fixed by the world market as at an established trade centre such as Philippines, Indonesia, India, Bangladesh etc. The price that can be paid on the island as a maximum is the world market price less freight. in reality, also less exporter margin. A US$ 450 per ton copra price in Manila, leads to a US$ 200 per ton FOB price in Honiara.
However, the problems do not end there. Copra is usually produced on a number of islands and is purchased on these to be moved to the shipping point which in the Solomon Islands is Honiara. There are two traders involved, The local one on the islands may be independent or merely an agent of the exporter trader in Honiara. There are also two shipments, one inter island and one international. Inter-island shipping tends to be expensive on a like for like mileage basis.
In overall terms, the price paid to farmers for coconuts ends up being very low. The only reason why they sell is that they have many coconuts and few alternative ways of securing a cash income. Not a very good way to make a living and bound to keep all in the chain poor. That is why there is not much competition to participate.