Our research on land rental markets and impacts of land registration and certification and land law reforms in Ethiopia (see www.steinholden.com/…) has shown that it is particularly poor female-headed households (landlords) that rent out their land to wealthier male-headed (tenant) households. This reverse tenancy system with poor landlords and less poor tenants has important policy implications. In our research we have found that land certification has strengthened the tenure security particularly of female-headed households and they have therefore become more willing to rent out their land and this has also enhanced the efficiency of land use as land is operated by more efficient land users. This is an example of a good policy that has had win-win effects improving efficiency and equity as we have found that the welfare improvement impact has been largest for female-headed households who get higher returns and higher food security from renting out their land (through sharecropping) than from trying to farm it themselves.

A new restriction on how much land to rent out was introduced in the new laws in Ethiopia in 2004-06, such that households were not allowed to rent out more than 50% of their land. We saw from our data that a large share of landlord households, and particularly poor female-headed households rented out more than 50% of their land. In our papers and feedback to policy makers in Ethiopia we warned against this new restriction and said it would be bad both for efficiency and equity and would particularly harm poor female-headed households. It is therefore now encouraging to see that here in Tigray Region, although this restriction is still stated in the law, it has not been enforced unlike some other elements of the law.

Another requirement in the new laws was that all land rental contracts should be registered at the tabia/kebelle (municipality) level, including all sharecropping contracts. In our surveys in 2006-07 we investigated household perceptions and opinions about this and based on this we recommended that this requirement should not be enforced but rather be given as an opportunity for households if they did not trust their contract parties. We were worried that this formalization requirement could increase the transaction costs in the land rental market and overburden the new local Land Administration Committees that are doing unpaid work and rather should focus their time on more important issues. It is therefore also good to see that this requirement is introduced in a soft way allowing that local bylaws are taken into account. In practice this appears to imply that sharecropping contracts among parties who trust each other are not registered and that is accepted. Again this is a good example of a rational and good adjustment of the policy and law enforcement.