Blog Image


Input Subsidies, Household Welfare, and Sustainable Land Use

This blog focuses on our recent research in Malawi on the Input Subsidy Program and Its Impacts. We have collected detailed panel data from farm households in 6 districts in central and southern Malawi in 2006, 2007 and 2009 and use these in our analyses. The research was funded by NORAD and the Norwegian Embassy in Malawi and data were collected by a group of students in our Master Program in Development and Natural Resource Economics

Towards more climate robust conservation agriculture in Malawi

Field work in Malawi Posted on Sun, September 05, 2010 20:38:28

The Achilles heel of the current agricultural policy is that maize is vulnerable to drought. We have analyzed data from three years with the input subsidy program where all the years were favored with good rainfall. Our data cannot therefore say so much about what the effects of a severe drought would be but the trends in the data over time may also illustrate a period of recovery after several years with unfavorable weather. The Ministry of Agriculture and Food Security now tries to reduce the vulnerability by expanding the extent of a conservation agriculture that is more robust to climatic variability. This involves use of more legumes, intercropping, organic manure, reduced tillage, herbicides, and agroforestry. Such changes in the production system should be stimulated at a broad scale as e.g. organic manure is still used by a small share of the households and many lack the knowledge of how to make organic manure from crop residues and green leaves. More use of Nitrogen-fixing crops and agroforestry trees may also reduce the need for importation of inorganic Nitrogen fertilizers and improve the soil quality. The ADP-SP program under ASWAp, among others, can contribute to the development of more sustainable, climate robust and cost-effective production systems that provide food security and cash incomes for the future generations. This should also reduce the need for imported inorganic Nitrogen fertilizers and also should reduce the risk of soil acidification due to excessive use of acidifying Nitrogen fertilizers.

How is the input subsidy program affecting investments in trees on farms?

Field work in Malawi Posted on Sun, September 05, 2010 20:29:22

Natural (indigenous) trees were found on 60% of all plots while exotic (planted) trees were found on 37% of all plots while the percentage of maize plots with natural and exotic trees were 66 and 44%. Exotic (planted) trees were more associated with household plots that had received subsidized fertilizer, possibly indicating a positive investment effect of fertilizer subsidies. There were also some signs that indicated that access to subsidized fertilizer could have stimulated more cutting down of natural trees (another investment effect). Livestock-rich households also had more plots with exotic trees while households with more real assets/capita were less likely to have natural trees on their plots. This could possibly also signal a transformation from natural capital (trees) to other assets. The probability that plots had natural trees decreased significantly over time, indicating a deforestation process. The probability that plots had exotic trees increased significantly from 2006 to 2009, possibly indicating that tree planting is on the increase, possibly due partly to the welfare improvement of households making them more able to invest. The probability of natural trees on the plots was much higher in Kasungu and Lilongwe districts than in the other districts and lowest in Thyolo and Chiradzulu while there was no significant difference between districts in terms of exotic trees.

Has the input subsidy program reduced the extent of intercropping?

Field work in Malawi Posted on Sun, September 05, 2010 20:25:30

While legumes were grown as the main crop on only 13% (2006) to 17% (2009) of the plots, legumes were the most frequently used intercrops. Intercropping took place on 54% of the maize plots versus only 15% of the plots under other crops. Pigeon pea was the most common intercrop found on 18% of the plots, followed by common beans (6% of the plots), cassava (5% of the plots). The fact that intercropping was more common on maize plots than for other main crops indicates that maize production does not necessarily imply mono-cropping of maize. A further inspection of the data revealed that 49% of the maize plots planted with hybrid maize were intercropped against 60% of the plots under local maize. When looking at fertilizer use versus intercropping, 46% of the maize plots that did not receive fertilizer had intercrops while 58% of the plots that received fertilizer had intercrops. Intercropping was much more common in Chiradzulu and Thyolo (82 and 78% of the maize plots) than in Lilongwe and Kasungu (18 and 29% of the plots). With smaller farm sizes in the south the maize area share increases but this is compensated by increased intercropping to enhance the production of these crops as well. Intensification therefore implies both more fertilizer use and more intercropping. No significant relationship between the extent of intercropping and access to subsidized fertilizer was found.

Has the input subsidy program lead to more mono-cropping of maize at the expense of other crops?

Field work in Malawi Posted on Sun, September 05, 2010 20:22:11

The assessment of crop choice at farm plot level revealed that 70% of the plots were allocated to maize in 2006 against 57% of the plots in 2009. Maize plots were, however, on average larger than the plots of all other crop categories. The number of plots under tobacco and sugarcane increased from 2006 to 2009. This may be associated with the introduction of subsidies for fertilizer for tobacco which partly also contributed to the overproduction of tobacco in 2008/09. There was also a positive trend in the percentage of the plots planted with legumes as the main crop, from 13% in 2006 to 17% in 2009.

Maize was found to be a more dominant crop on smaller farms. The average maize area was 0.71 ha/farm and the average farm size was 1.17 ha. An increase in the farm area of 1ha was associated with an increase in the maize area of 0.48 ha. Labor- and livestock-rich households have significantly larger maize areas. The maize area has reduced significantly from 2006 to 2009 showing that maize production is intensified as fertilizer use has increased in the same period. The result of increased input access and use is therefore area intensification rather than area expansion. This may imply that the subsidy program also facilitates production of other crops by releasing maize areas.

The mean maize area share out of total farm size is 0.68, varying from 0.80 in Thyolo to 0.57 in Kasungu and Machinga districts. The maize area share has decreased from 0.73 in 2006 to 0.64 in 2009. The land-poor have larger maize area shares while the livestock-poor have smaller maize area shares.

Impact of access to subsidized fertilizer on maize yields: Were households that received subsidized fertilizer able to use it efficiently?

Field work in Malawi Posted on Sun, September 05, 2010 20:17:52

Access to subsidized fertilizer had a significant positive effect on maize yields. However, the targeted households had significantly lower maize yields than those not targeted by the program, whether receiving subsidies or not. This was found by assessing the yields of households that had been erroneously excluded and included in the program based on our participation predictions. This may indicate that the subsidies have been systematically targeted towards less efficient farm households.

A closer inspection of how asset poverty is related to maize productivity revealed that the relative land-poor (bottom half in land/capita) had maize yields that were 360-380 kg/ha higher than the relatively land-rich households. On the other hand, the relatively labor-poor households had on average maize yields that were about 360 kg/ha lower than that of the relatively more labor-rich households.

Maize yield variation in different districts in Malawi from 2006-2009

Field work in Malawi Posted on Sun, September 05, 2010 20:08:01

Maize yields were also significantly higher in Thyolo (2590 kg/ha as average over the three years) than in other districts. Maize yields were particularly skewed in Machinga district (mean yield of 1400 kg/ha against a median yield of only 457 kg/ha). Hybrid maize yields were found to be significantly higher (about 320 kg/ha higher) than the yields of local maize also after controlling for differences in fertilizer and manure use (using propensity score matching). There was a significant positive trend in maize yields from 2006 to 2009 with an increase in mean yields of about 600 kg/ha from 1440 to 2040 kg/ha for hybrid maize and from 1120 to 1680 kg/ha for local maize. These findings illustrate that there is still a lot of room for yield improvement, particularly for hybrid maize. Only in Thyolo district were fertilizer use levels at the recommended levels on average (345 kg/ha) while the median fertilizer use level was 200 kg/ha. In the other districts mean fertilizer levels were 140-210 kg/ha and median levels were 70-125 kg/ha with more than 25% of the maize plots receiving no fertilizer in Chiradzulu, Machinga, Kasungu and Lilongwe districts. The mean fertilizer levels increased from 193 kg/ha in 2006 to 211 kg/ha in 2009 on all maize plots and the distribution became less skewed over time with more than 75% of the plots receiving fertilizers in 2009 and a median fertilizer rate of 151 kg/ha in 2009 against 107 and 64 kg/ha in 2007 and 2006.

Does the introduction of cheap fertilizers crowd out organic manure?

Field work in Malawi Posted on Sun, September 05, 2010 20:05:10

For the decisions to use fertilizer and manure on plots it was found that these inputs are mostly used as complementary inputs and not as substitutes when we analyze all plots together. When analyzing maize plots alone, the use of fertilizer and manure were not significantly associated. There seems therefore to be little reason to fear that input subsidies crowd out the use of manure. The main problem is that the use of manure is limited as a large share of the plots did not receive manure. However, use of organic manure is expanding, probably due to the extension efforts and increasing emphasis on conservation agriculture methods also linked to the fertilizer subsidy program. Households that did not receive subsidized fertilizers were less likely to use fertilizer on their plots but there was no effect on manure application on maize plots.

The intensity of use of fertilizer and manure was also positively correlated when analyzing all plots but not so for maize plots. Access to subsidized fertilizer enhanced fertilizer use intensity on maize plots. Use of hybrid maize was positively associated with higher fertilizer use intensity as well as manure use intensity. Both fertilizer use intensity and manure use intensity increased from 2006 to 2009. Fertilizer use intensities were significantly higher in Thyolo and Zomba districts than in the other districts.

Impacts of input subsidies on the agricultural cropping system in Malawi

Field work in Malawi Posted on Sun, September 05, 2010 19:45:44

We have assessed the agricultural cropping system in central and southern Malawi based on farm plot level data from more than 400 rural households and more than 4000 farm plot observations in six districts covering the years 2006, 2007 and 2009. In particular the study has attempted to identify effects of the Malawian input subsidy program (FISP) on the cropping systems. The types of effects looked at include;

· use of fertilizer and organic manure (whether these inputs are used or not at farm plot level),

· intensity of use of fertilizer and manure,

· the use of alternative maize varieties (hybrid, open-pollinated, and local),

· the productivity differentials between the maize varieties,

· input use and maize productivity,

· crop choice (choice between maize, legumes, root and tubers, other cereals, and tobacco/sugarcane

· factors affecting the household maize area and maize area share of total farm size

· extent of intercropping, crops used for intercropping, and decisions to intercrop at farm plot level

· presence of natural (indigenous) and exotic (planted) trees on plots.

The cross-cutting issues were:

· How has access to input subsidies for fertilizer affected these variables?

· How are they affected by asset poverty?

· What are the trends over time and variation across districts in Malawi?