Effect of land tenure on forest cover and the paradox of private titling in Panama

Highlights

• Protected Areas and indigenous Comarcas decreased 1990–2020 deforestation in Panama.

• Private titling campaigns increased deforestation in Panama from 2000 to 2011.

• Private titling increased deforestation during titling and years after in Panama.

• Titling as a precondition for conservation programs like REDD+ may cause forest loss.

• Environmental markets for reforestation can increase speculative deforestation.

Abstract

Meeting sustainable development goals requires policies that account for interrelatedness in social and environmental issues such as land tenure and deforestation. This work takes advantage of a nationwide titling campaign in Panama to explore the effect of private titling on forest cover across a heterogeneous landscape covering all stages of forest transition and diverse tenure arrangements. Situated in a broader matched analysis of the influence of zoning and tenure on forest cover, private management is estimated to have contributed to the deforestation of 1750–3650 km² of mature forest nationwide from 1990 to 2020 with an average marginal effect of 15.3%. Conversely, Protected Areas and Indigenous Comarcas are estimated to have protected 1700–3900 km² and 500–1250 km² of mature forest, respectively. Private titling is associated with increased deforestation both during titling and years after, supporting observations that the titling process itself encourages speculative deforestation by title seekers and that private landholders value natural forests less than other land uses such as cattle. By disaggregating the data by region to highlight different stages of forest transition as well as by processes of deforestation and forest growth, this analysis shows that while private titling accelerates deforestation, it also encourages investment in reforestation. This presents a paradox for private titles and forests where agencies may perversely encourage speculative deforestation by creating stronger markets for forest-ready landscapes than for intact natural forests. In cases such as this one, where deforestation helps to secure a title, this paradox is confounded when having a title is set as a precondition for participation in a forest conservation program.

Introduction

Tackling global sustainability challenges requires solutions that simultaneously address interrelated components of socio-ecological systems (Berkes et al., 2003, Folke et al., 2002, Sayer et al., 2013). Forests play a prominent role in sustainable development through critical ecosystem services, climate change mitigation and adaptation, biodiversity protection and human livelihood security (Katila et al., 2020, Timko et al., 2018, Seymour and Busch, 2016). Despite a growing collection of international agreements to conserve forests, however, deforestation has continued at an unsustainable pace (Baccini et al., 2017, Curtis et al., 2018, Nydf Assessment Partners, 2019; IPBES, 2019; WWF, 2018). Recent expansion of planted forests, primarily in temperate zones (Köhl et al., 2015), has caused a decline in net forest loss (Song et al., 2018, Fao, 2015), which may allow global forest trends to be framed in the optimistic terms of a forest transition (Song et al., 2018, Rudel et al., 2019, Meyfroidt and Lambin, 2011). Implications for sustainability and livelihoods, however, require nuanced examination of pathways of avoided deforestation and forest growth (Griscom et al., 2020, Naudts et al., 2016; WWF, 2018) as well as regional asymmetry in forest trends.

Curbing tropical deforestation is essential to achieving climate stability (Seymour and Busch, 2016) and protecting biodiversity (Bradshaw et al., 2009). However, due to large numbers of rural poor in forest-rich tropical countries (Wunder, 2001, Sunderlin et al., 2008) and the highest global rates of urbanization and development (Swamy et al., 2018), policy prescriptions for achieving sustainable development goals in the tropics often pit people against forests (Chomitz, 2007, Hartshorn, 1995). This is commonly the case in policies relating to tenure, or the institutions concerning who can access and benefit from resources (FAO, 2002). Tenure policies favoring forest conservation often exclude peoples’ access to forests through protected areas (PAs). While PAs are generally effective in reducing deforestation (Busch and Ferretti-Gallon, 2017, Min-Venditti et al., 2017), to the point that they collectively reduced tropical carbon-based emissions by around 30% from 2000 to 2012 (Bebber and Butt, 2017), their exclusionary nature often negatively impacts the livelihoods of people around them (Oldekop et al., 2015). Alternative tenure arrangements such as communal management and private titling allow for people to garner economic benefits from forested lands, but have more varied success in conserving forests.

The literature on the effect of tenure arrangements on forest conservation is deep and varied, yet it usually relates to communal forest use and management. While not a panacea (Baynes et al., 2015, Ostrom and Cox, 2010, Holland et al., 2017), communal management is generally found to benefit forest cover (Porter-Bolland et al., 2012, Min-Venditti et al., 2017), especially when the community is homogeneous with low immigration and high autonomy (Poteete and Ostrom, 2004, Agrawal and Chhatre, 2006, Chhatre and Agrawal, 2009). Since the mid-1980s, land tenure reform has dominated many international conservation and development policies. While such policies have focused largely on strengthening communal tenure arrangements, there are many situations where communities do not fit conservation-friendly prototypes or where poor land users do not belong to such communities. In these spaces, issues of individual tenure have been the focus of these policies in a simultaneous wave of private land-titling campaigns across the tropics. The area of tropical forest owned privately increased by 122% from 2002 to 2008 (ITTO, 2009). While similar to earlier land reforms in their primary goal of poverty reduction, these titling campaigns have differed in their general claim to underpin broader agendas of sustainability and forest conservation (Pacheco et al., 2011, Sunderlin, 2011).

Securing land and property rights is considered critical to achieving the Sustainable Development Goals (IEG, 2016) as a tool for reducing poverty and enhancing economic development. Development institutions that advocate for and facilitate private-titling campaigns generally maintain that private titling is likely to have a positive or neutral effect on forests (Keipi, 1995, Fao, 2012, Deininger, 2003), although long term effects remain understudied (Lawlor et al., 2020). Conservation literature suggests a more dubious relationship between private titling and forest conservation even in the short term, however. Recent meta-analyses of the effect of private land tenure on forest cover have found mixed results (Min-Venditti et al., 2017, Katila et al., 2020, Busch and Ferretti-Gallon, 2017). Clear ownership of forests can facilitate use of market mechanisms to manipulate incentives toward forest conservation or afforestation and is thus often a prerequisite for programs such as REDD+ and other Payment for Ecological Service (PES) programs. Whether such projects are successful depends both on context-specific cultural and economic factors as well as on the process of forest-cover change in question.

In cases where forests have already been degraded or deforested, secure private tenure can lead to greater forest cover if it provides landholders with incentives or means to invest in agroforestry or silviculture (Besley, 1995, Takahashi and Otsuka, 2016) or to intensify agricultural production and thus spare other lands for regeneration. Conversely, tree planting can be a means to demonstrate investment and thus enhance tenure security (Barbier and Tesfaw, 2013, Sjaastad and Bromley, 1997). If titled, degraded lands that no longer offer profitable agricultural returns can also be allowed to regenerate without fear of being taken by squatters or the government due to inactivity (Kaimowitz, 1996).

The mechanisms through which private titling might curb deforestation in existing forests are less straightforward. Secure private tenure can theoretically enhance forest conservation if it obviates motives to cut trees to prevent others from doing so first (Kaimowitz, 1996) or empowers landholders to prevent others from clearing forests (Alston et al., 2000). However, these mechanisms only work if landholders consider forests more valuable than other land uses, which is often not the case (Liscow, 2013, Angelsen and Kaimowitz, 1999, Angelsen, 2007, Robinson et al., 2017). Formal titles may also allow speculators to leave land idle without fear of invasion (Alston et al., 1996, Azevedo et al., 2017). However, in the context of development, such land speculation is generally not beneficial to poor smallholders (Fairhead et al., 2012). Largely for this reason, many governments have historically recognized claims to lands only when landholders can prove use, usually via clearing. When land clearing enhances one’s claim to the land, private titling has a negative relationship with forest cover (Angelsen, 2007, Arnot et al., 2011, Araujo et al., 2009). One of the main reasons for titling private land advocated by development agencies is that such titles enhance landholders’ access to credit and ability to invest (Dorner, 1972, Feder and Feeny, 1991, IEG, 2016). However, once provided the means, landholders may invest in deforestation (Rasmussen et al., 2017, Deininger and Minten, 1996). By entering forested lands into the market economy via formal titles, it is also more likely that they will eventually be sold to larger landholders such as ranchers (Schneider, 1994, Campbell, 2015). Empirical studies of efforts to establish or clarify private tenure in forested landscapes are rare but suggest it is risky for both people and forests (Robinson et al., 2014).

There have been few studies on the effect of private tenure on forest cover in Central America in the last decades (Min-Venditti et al., 2017) despite several large-scale land titling projects implemented in the same period (Keipi, 1999, Deere and Leon, 2002). From 2001–2010, the moist forests of Central America suffered net forest loss, although, as with global forests, this trend can be classified as an asymmetrical forest transition (Redo et al., 2012) as losses were partially offset by net gains in dry and coniferous forests. In Central America, land tenure is often secured by clearing forest (Ankersen and Ruppert, 2006, Liscow, 2013, Angelsen and Kaimowitz, 1999, Jones, 1990). Land speculation by cattle ranchers is considered a principal cause of deforestation in Latin America (Roebeling and Hendrix, 2010), and there is evidence that recent land titling campaigns have fueled this speculative drive for land (Kaimowitz, 1996). While land policies and private tenure likely influence deforestation, they also influence forest recovery (Pacheco et al. (2011)

This work explores the effect of tenure on forest cover in the Central American nation of Panama and takes advantage of data from a large-scale private titling campaign to elucidate the effect of private tenure and titling on forest cover. Although a small nation of around 75,500 km², Panama presents an interesting microcosm of Central America with its diverse representation of land uses and tenure arrangements and simultaneous presence of all three forest transition stages, with “settled”, “frontier” and “remote” zones (Perz and Skole, 2003). Although some consider Panama to have already undergone a forest transition due to regrowth of forest in the settled region (Redo et al., 2012, Wright and Samaniego, 2008, Hosonuma et al., 2012, Sloan, 2015), deforestation has continued at a steady rate since 1990 in other regions (Walker, 2020). The national extent of the land titling campaign and other tenure arrangements amidst this mosaic of forest processes allows for insights into the effect of tenure and titling on forests to be broken down by processes of deforestation and forest growth. Such disaggregation is important to elucidate true impacts on forests and consequences for biodiversity and climate change mitigation.

The explicit goal of the titling program was to reduce rural poverty by increasing farmers’ access to credit (IDB, 2014). A full review of whether this formal privatization of land and its insertion into the global market has had the intended effect of poverty reduction in Panama is beyond the scope of this paper (see Spalding, 2017). Here I focus on the environmental impact of titling and specifically on its effect on forest cover. Despite publishing a lucid report on the tenuous and often catastrophic relationship between private titling and forest cover in Central America (Jaramillo and Kelly, 1999) along with a long passage describing the history of deforestation to gain possession of land in the principal loan proposal itself (IDB, 2002), the InterAmerican Development Bank concluded in various loan documents that the titling program was expected to have positive or neutral effect on the environment. This work explores that relationship.

Exploration of effect of private titling on forest cover is first situated in a broader analysis of the influence of different zoning/tenure arrangements on forest cover in Panama. In assessing policy decisions such as zoning and titling on deforestation, the counterfactual, or what would have occurred in the absence of a given tenure arrangement, cannot be observed directly, and can only be approximated by controlling for environmental and social variables that influence treatment and outcome (Burivalova et al., 2019, Ferraro, 2009, Blackman, 2013). I controlled for such endogenous factors with propensity score matching, which balances the treatment and control groups by equalizing the probability of treatment based on a set of observed factors (Rosenbaum and Rubin, 1983). I first estimated the effect of two restricted zones, PAs and Indigenous Comarcas, on deforestation from 1990 to 2020 across Panama and compared these to estimates for effect of private management and other Indigenous territories (where data allow) on deforestation using the same methods. I then examined the effect of private titles explicitly by asking (1) whether the PRONAT land-titling campaigns favored parcels with less forest and/or increased recent deforestation rates, and (2) whether deforestation or regeneration rates changed following attainment of a private title. Due to the relatively short amount of time since titling for many of the titles issued during the PRONAT campaign, I repeated this latter analysis for titles granted prior to the PRONAT campaign.