survey Tanzania
Photo credit: Aurélie Harou

Tanzania Field Trial Finds On-site Soil Testing and Subsidies Can Increase Fertilizer Use and Maize Yields

The right mineral fertilizers applied appropriately can alleviate nutrient deficiencies in soils, but most small-scale farmers in Sub-Saharan Africa do not have their soils tested to reveal these deficiencies.

A multidisciplinary research team supported by USAID just published results from Tanzania showing that targeted fertilizer recommendations from low-cost, on-site soil tests paired with subsidies to purchase the recommended fertilizer can increase maize yields on small-scale farms. These soil tests also showed that a change in basic national fertilizer recommendations provided by the government of Tanzania could address a common, important soil deficiency in Morogoro, where the study took place.

“If you can produce more crops, you have more to eat,” said Cheryl Palm, the soil scientist at the University of Florida who led the research. “You might produce enough to sell, and that’s very often what happens.”

An Onsite Soil Testing Tool for Targeted Recommendations

Most small-scale farmers in Tanzania and in much of Sub-Saharan Africa invest very little in chemical fertilizers. One reason could be that the fertilizers recommended by the national extension service do not address nutrient deficiencies in their soils and therefore have only limited impact on yields.

Like other national extension services, Tanzania’s provides recommendations for the type and amounts of chemical fertilizers to apply for various crops. Government recommendations for maize cover entire regions, encompassing broad areas with considerable differences in soils and growing conditions. Uniform recommendations may not reflect nutrient deficiencies on an individual farmer’s plot, limiting the effects of the fertilizers on crop yields.


SoilDoc: Bringing the Lab to the Field

SoilDoc is a portable, on-farm soil testing kit that would provide tailored fertilizer and organic input recommendations for farms, following the Integrated Soil Fertility Management (ISFM) approach. SoilDoc diagnoses soil constraints in the field, transmits the information quickly through SMS with an android phone or tablet, and sends quality recommendations on time. In this way, farmers are advised on best management of their farm to increase production and apply costly fertilizers only where needed. SoilDoc was developed through a collaboration of researchers at the University of Maryland and Columbia University led by soil scientist Ray WeilLearn more at the Columbia University Earth Institute.

Researchers at Columbia University’s Earth Institute developed SoilDoc, a low-cost on-site soil testing tool that quickly identifies nutrient deficiencies and provides targeted fertilizer recommendations. SoilDoc was first developed with the support of the Alliance for a Green Revolution in Africa (AGRA).

SoilDoc testing and recommendations cost about 7,000 TZS per plot (about $3 USD). The tests can be carried out by government agricultural extension agents and are simple and fast enough that a single agent can carry out tests on multiple plots during a single trip.

Testing the Impact of Information on Maize Yields in Tanzania

Starting in 2013, Palm led a multidisciplinary team of soil scientists, agronomists and agricultural economists from Columbia University, McGill University, the University of Florida, the University of Illinois Urbana-Champaign and Sokoine University in Tanzania. The team conducted a randomized controlled trial (RCT) to test the impacts of plot-level soil tests and targeted recommendations using SoilDoc. The study took place in Morogoro, Tanzania, an area of good agricultural potential but low maize yields. About 40 percent of households in the area have incomes below the $1.90/day (2011 PPP) poverty line.

The 1,007 plots sampled in the study were all rainfed with no irrigation. When the study began, less than one percent of farmers in the sample used any chemical fertilizers on their maize crops.

For the RCT, farmers from 50 villages were randomly assigned to one of four groups:

  • Farmers received the SoilDoc fertilizer recommendations for their plot with a voucher they could take to an agrodealer and redeem for chemical fertilizer or other agricultural inputs, cash or a combination of the two.
  • Farmers received the same voucher but no recommendations.
  • Farmers received recommendations but no voucher.
  • Farmers received neither recommendations nor a voucher so the research team could compare their maize yields and fertilizer purchases against the other three groups.
soil testing in Tanzania

The project tested soils on all main maize plots. At the end of the study, the research team provided the SoilDoc testing results and fertilizer recommendations to all farmers in the study.

To make sure that the recommended chemical fertilizers were available, the research team guaranteed agrodealers willing to make the drive a minimum level of sales. The majority of these agrodealers had very small operations. They agreed to spend six hours in each village to offer all chemical fertilizers recommended by SoilDoc as well as hybrid seeds and other inputs at the same prices to all farmers.

“The agrodealers showed up, and all farmers involved in the study got a phone call for notification,” said Hope Michelson, an agricultural economist at the University of Illinois who took part in the study. “We addressed the last mile problem for everyone in the study. That’s important to understand.”

The Limits of Information for Driving Maize Yields

A new paper published in the Journal of Development Economics with lead author Aurélie Harou, an assistant professor of agricultural and resource economics at McGill University, reports the team’s findings.

The initial idea behind SoilDoc was that information on its own would be empowering and transformative for farmers. However, only farmers who received both the SoilDoc recommendations and the subsidy increased their use of chemical fertilizers. They were also the only group of farmers whose yields increased.

“We don’t see evidence that people act on these personalized soil recommendations without a subsidy. Not yet anyway,” said Michelson.

The team quantified the average value of additional yields generated by the SoilDoc recommendations combined with the fertilizer subsidy. The recommendations together with a subsidy increased farmers' profits by an average of 37,000 Tanzanian shillings (about $16) on a one-acre farm, even if farmers pay the full costs of soil tests. The increase is equivalent to about seven days of work at the median daily wage.

“The effect of providing soil testing and vouchers is striking because it’s hard to get yields to move. Yields are prone to weather-driven shocks that add a lot of statistical noise in a study like this. It can be difficult for that reason to detect changes,” said Michelson.

She added, “The result stands out among similar studies because of the magnitude of the yield change and its statistical significance. We see behavioral change in terms of fertilizer purchasing and application increases in response to providing vouchers and information, and that behavioral change results in higher yields.”

Michelson also said that weather-related risk may have had an impact on how much farmers were willing to invest in chemical fertilizers to increase yields. According to USAID, Tanzania has had six major droughts over the past 30 years, and rainfall patterns have become increasingly variable since the 1960s. Chemical fertilizers, like any input that comes with an extra cost, increases how much a farmer will lose if a crop fails due to drought.

“In this case it could be related to uninsured production risk as well as insufficient liquidity to buy fertilizer,” said Michelson.

In 2016, a drought caused steep drops in maize yields across the region where the study was taking place. However, farmers who applied fertilizer had less of a decline in yields.

Improving National Fertilizer Recommendations in Tanzania

The SoilDoc tests showed that nearly every plot included in the study was deficient in sulfur. Sulfur is critical for maize productivity. Plants need sulfur to produce proteins and chlorophyll for healthy growth. A sulfur deficiency causes a maize plant’s new leaves to turn yellow. As the plant continues to grow, the deficiency can lead to significantly lower yields.

Sulfur is not included in current regional or national fertilizer recommendations from the Government of Tanzania.

“The correct fertilizers are not even available to people in many of the rural areas,” said Palm. “The fertilizers they do have are in general not the ones that are needed to get the best crop response.”

Michelson said that fertilizers with sulfur are more common in Tanzania today than they were in 2013 at the beginning of the study. She cited that the fertilizer company Yara Tanzania has introduced a urea fertilizer that includes sulfur.

Palm said that an important context for efforts to increase small-scale farmers’ use of chemical fertilizers is that it isn’t going to destroy the environment or poison the soil.

“The amounts of fertilizer being used by farmers in our study are quite small,” said Palm. “So the chances of leaching and other types of environmental contamination are also very small. At their current application rates, greenhouse gases and leaching do not increase at all.”

According to World Bank data, farmers in Tanzania applied an average of 15.9 kg/ha of chemical fertilizers in 2018, compared to 128.8 kg/ha applied by farmers in the United States. At the same time, that number in Tanzania is a substantial increase from below 1 kg/ha in 2001.

“These are not intractable problems,” said Palm. “There was just a long time without updated research and agricultural extension that are needed to increase agricultural productivity. A lot of effort is being put into it now.”

Read the paper in the Journal of Development Economics: “The joint effects of information and financing constraints on technology adoption: Evidence from a field experiment in rural Tanzania,” by Aurélie P. Harou, Malgosia Madajewicz, Hope Michelson, Cheryl A. Palm, Nyambilila Amurie, Christopher Magomba, Johnson M. Semoka, Kevin Tschirhart and RayWeil.