RE: What can we do to improve food security data? | Eval Forward

I think this is an excellent example of the consequences of over extending the financial resources of a host country. As I look at the overall economic conditions that are common to and effectively define developing countries, it is they represent financially suppressed economies. By this I mean economies that serve a generally impoverished population in which the bulk of the populations spend up to 80% of income or farm production on basic subsistence needs such as marginal food, clothing and shelter. While this can put tremendous downward pressure on consumer prices, there is also no discretionary funds to form a tax base that will fund civil services such as statistical analysis. Sorry, but no taxes; no services. Thus, governments are for all practically purposes financially stalled, barely meeting personnel financial obligations of salary plus benefits with virtually no operating funds for implementing programs, such as statistical surveys. This results in surveys being filled in by the enumerators in accordance with their biases or expected survey biases.  However, the results are accepted as accurate by the government with policy decisions being based on the results. Governments then heavily rely on external donor funded NGO to do much of the development activities. Such NGO often have their own objectives or are contracted to impose donor motivated solutions that may or may not be consistent with the needs of the beneficiaries such as smallholder farmers.  Please review the following webpages and links: https://smallholderagriculture.agsci.colostate.edu/financially-suppressed-economy-2/ and https://smallholderagriculture.agsci.colostate.edu/financially-stalled-governments/

Looking at the specific Benin survey case, I sense that being somewhat aware of the budget limits the organizers tried to compensate by making the survey more complex and more demanding on the enumerators. As an agronomist my discipline does not normally rely on survey information and are not fully familiar with survey technology. However, I am somewhat unusual for an agronomist and have often relied on survey information to get a fuller understanding of farming systems. Thus, I am familiar with some of the guidelines for conducting a quality survey. My normal guidelines are that a survey should take no more than an hour, an enumerator should conduct only 4 or at most 5 interviews per day. Thus, expecting them to do a 15-page survey would be very difficult in an hour and 10 per day virtually impossible. Thus, the enumerators will have to resort to group interviews that I tend to dismiss as having individual biases of outspoken minorities in them, or simply fill out the surveys on behalf of the interviewees. If you are expecting results from remote areas and not able to provide transportation, is there an alternative to having the enumerators fill out the surveys and still meet the limited deadlines for completing the survey? I don’t think so.

The result is the data is full of bias and unreliable, but consistent with the financial state of the country. It also means that the government is substantially out of touch with most of the populations. I recall hearing a few years ago that about 90% of smallholder farmers have never had any contact with agriculture officials and they are perhaps better off because of that.

A more specific case that I am personally knowledgeable of would be the Ruaha river basin in Southern Tanzania.  Here there were many complaints that the river was drying up earlier each year and endangering the wild life in the Ruaha National Park, which was blamed on climate change. According to the government there were 4 rice schemes above the National Park totally some 11,000 ha. However, when they got an aerial or satellite view the total rice acreage was some 40,000 ha. The difference was many small illegal informal schemes build by the farmers even including some permanent concrete abutments to divert the water.  Now what civil officers is going to wonder through the area on a rough road that will take at least 6 hours to fully transvers the 200 km and tell the poor smallholder farmers that must stop growing rice so foreign visitors can enjoy the wild life in the National Park. More likely they will visit the area, inform the farmers the rice cultivation is illegal, but with a modest gratuity allow them to continue until they return in 3 or 4 years. What government official will acknowledge this? Easier to simply continue to blame the river drying on climate change.

So, what can be done to get more reliable survey information? I don’t really know. I would suggest extrapolating from donor assisted NGO administered projects where they do have enough budget to manage a reliable survey. This would make for a patchwork of quality information across the country, but I would expect it to be fairly representative. Until the economy can expand to provide a reasonable tax base to finance civil services I fear this is the best means forward.

I hope this is helpful.

Thank you.

Dick Tinsley