A lot of us who may come from the West assume that land rights
certification, registration or titling are important attributes of any
kind of land tenure or property rights system. We think of formal
recording of land rights as essential to assuring farmers that they have
land tenure security, an important enabling condition to agricultural
development.
Economic theory and common sense tell us that if a family is going to
invest in their property, they need to have a clear expectation that,
far into the future, the kinds of sacrifices, investments of labor,
capital, materials, into that land, and the benefits that come from
those investments, will accrue to them. There’s a very simple
relationship between land tenure security, property rights security, and
investment: Theory predicts positive outcomes, and these are often
observed practice, where people have clear tenure security.
However, in many developing countries, the kind of formal
certification, property rights and titling systems that we are familiar
with in wealthy countries often do not exist. A lot of farmers farm on
land owned by the state. In Africa particularly, a lot of farming—up to
90 percent—is done on land held under customary tenure regimes, where
land rights are not certified formally. Under customary tenure, people
gain access to land as a social right, granted by virtue of their
membership in a community.
THE QUESTION
So in 2013, the UK Department for International Development (DFID)
asked me and a group of scholars to pull together a team of other
researchers and graduate students to do a systematic review on the
effects of efforts to “formalize” land rights through certification or
titling on changes in agricultural investment and productivity in the
developing world. Theory would predict that where farm land was
previously unregistered, levels of investment and productivity would
increase markedly after certification or titling.
Part of the development discourse over the past 30 to 40 years has
been that African agriculture will not take off unless people have clear
tenure security, and there’s an underlying assumption that this is
delivered through land-rights certification or titling, as in Latin
America and parts of Asia. So, in light of the fact that over the past
30 to 40 years there have been a number of efforts to convert
non-formally tenured regimes into formally tenured regimes, based on
certification by the state, the question that DFID asked us to look into
was, What have been the effects with respect to the expected increases
following land-rights certification and investment in productivity? In
farmer incomes? In flow of credit?
That was the topic of our systematic review.
Our inclusion criteria was limited to studies that were based on
randomized control trials—randomized samples of farming households in an
area that had received a treatment, which was, for example, land-rights
certification, in comparison to a community where we were able to
control all other factors apart from the fact that the community had not
received land-rights certification. We wanted to look at, empirically,
the effects of the certification intervention on investment,
productivity of agriculture, farmer family incomes, and access to
credit.
THE FINDINGS
Those who have conducted systematic reviews know the pain. The
graduate students looked at 25,000 titles on this subject and reduced
that to a review of 1,000 abstracts, which yielded 100 papers that we
looked at in detail. Only 20 studies met our inclusion criteria. This is
a huge question for the development world, for economic theory, a whole
host of issues—and only 20 studies met a rigorous standard of empirical
research design.
The 20 studies fell in nine countries: five in Latin America, five in Asia, and 10 in Africa.
In the Latin American and Asia cases, after certification or titling,
there were significant gains to productivity of between 50 and 100
percent, and strongly positive gains to investment and income following
tenure recognition, typically titling. However, in the Africa cases,
there were weak or modest gains to productivity—between zero and 10
percent gains to productivity—and in investment and income following
certification (though in most cases there were still positive gains)
Another important finding was there was no or weak discernible credit
effects anywhere. Most studies—and we were looking very carefully
through a gender lens with respect to differential effects on men and
women—failed to aggregate effects of tenure recognition on women, except
for two quantitative studies that identified positive effects (in
Ethiopia and in Rwanda).
WHY DID RESULTS DIFFER?
So then the question becomes, Why are these significant gains in
Latin America and Asia, and these relatively weak or modest gains in
Africa? We have three hypotheses that we’re exploring through further
research.
The first hypothesis is what we’re calling the role of pre-existing
institutions—in Africa specifically, customary tenure. Customary tenure
systems provide access to land as a social right by virtue of one’s
membership in the community. An indicator of the security of the tenure
is that land can often be inherited by other family members, but it
can’t be sold, typically. Customary tenure often provides high levels of
tenure security.
Customary tenure systems generally provide poor people in Africa
access to land, free of charge, and once again as a social right; this
is a pervasive institution in Africa. The designers of certification and
titling programs, we hypothesized, were likely underestimating the
tenure security of people who held those lands. And so when those land
rights were certified, the kind of productivity gains or investment
gains that would have been projected, assuming prior tenure insecurity,
didn’t happen. Those assumptions were misplaced.
Another factor is what we’re calling the wealth effect: Household
resources and income in Africa are much lower among poor farmers in
comparison to poor farmers in Latin America and in Asia. So, if you’re
going to do something with your land, it’s just not about land as an
asset. It’s about labor, capital, having the income to invest in your
farming enterprise, and the generally low levels of income among African
farmers constrain their ability to make better use of their land.
That’s hypothesis number two.
The third hypothesis is what we’re calling the effects of
complementary public investments. Programs to secure land rights are
best treated as one element in comprehensive agrarian reform programs.
Effective reform is not about providing secure land rights to people.
It’s also about providing affordable access to farming inputs and
markets, and investment in roads, cooperatives, farming training, and so
on; investments that enable farmers to capitalize on their secure land
rights. Levels of public investment in rural areas in Africa are, we
believe, much lower than they are in Latin America and Asia.
One of our arguments is that, when talking about land-rights
certification or formalization in Africa, you really have to approach it
as a package of investments, and you have to account for this wealth
effect.
Our plan is to explore more deeply what might explain the weaker
responses in Africa. We believe the hypotheses I’ve noted provide a good
starting point. BY STEVEN LAWRY.