Land governance as a precondition for decreasing deforestation in the Brazilian Amazon

Land Use Policy xxx (xxxx) xxxx

Contents lists available at ScienceDirect

Land Use Policy journal homepage: www.elsevier.com/locate/landusepol

Land governance as a precondition for decreasing deforestation in the Brazilian Amazon Bastiaan Philip Reydona, Vitor Bukvar Fernandesa,*, Tiago Santos Tellesb a b

Institute of Economics, University of Campinas, Rua Pitágoras 353, CEP 13083-857, Campinas, São Paulo, Brazil Agricultural Research Institute of Paraná State, C.P. 10.030, CEP 86057-970, Londrina, Paraná, Brazil

ARTICLE INFO

ABSTRACT

Keywords: Deforestation Property rights Amazon rainforest Land policies Land governance

This paper’s aim is to show that improvements in land governance in Brazil, and particularly in the Amazon region, have been the main pre-condition enabling reductions in deforestation of the Amazon rainforest. Deforestation occurs primarily where property rights are not clearly established, and occurs mostly on land directly or indirectly under state responsibility. This paper also shows that land speculation plays an important role in deforestation. Based on these findings, it is evident that Brazil must improve its land governance in order to decrease rates of deforestation. The Brazilian government has adopted important new efforts to improve land administration, including improvements in the registration process and the Terra Legal program, which are addressing public land problems across large areas of the Amazon. The concluding section highlights how efforts to reduce deforestation will only be possible through more efficient land governance, especially in the Amazon region, and discusses the important role that participatory land governance may play in improving land use and land ownership through application of land taxes.

1. Introduction While the fact that Amazon deforestation is a pressing matter that needs to be dealt with is true, the ways to solve this issue and the diagnosed causes of the deforestation are subjects of heated debate. These debates most often revolve around changes in the Brazilian Forest Code (Sparovek et al., 2012; Stickler et al., 2013; Soares-Filho et al., 2014) and the recent increase in rates of deforestation (Arima et al., 2011; Nepstad et al., 2014; Gibbs et al., 2015; Pfaff et al., 2015; Sonter et al., 2017). After 2015, Brazil is going through an ongoing political and economic crisis that has resulted in ever larger budget cuts, especially in science agencies. Since early this year the recent-elected Bolsonaro government is continuing the overall budget cuts and also favoring agroindustry and mining interests that intensify these activities in

Amazonia (Artaxo, 2019). To better understand these phenomena, it is necessary to clearly identify the deforestation process, determine the causes, reflect on short, medium, and long-term solutions. It is undeniable that the strong command and control policies1 and economic incentives2 implemented from 2004 onwards have played a crucial role in reducing deforestation (Lapola et al., 2014). However, since these policies depend on the direct intervention of the State, it is difficult for them to be maintained over the long term. This is mainly because the principal agents responsible for deforestation, including livestock ranchers to grain producers and electricity generators, are well-established and will persist in the region, requiring more lasting solutions to be found. The core aim of the present study is to build a strong argument showing that the pre-condition to solving Brazilian land problems lies with the State, in conjunction with civil society,

Corresponding author. E-mail address: [email protected] (V.B. Fernandes). 1 The main government command and control policies focused on changing the behavior of deforesters, were: (a) the federal police operations known as Curupira in 2005 and Arco de Fogo in 2008, to combat illegal timber extraction; (b) Decree 6321/2007, which restricts the granting of bank loans and obliges owners in municipalities that are the biggest deforesters to reregister; (c) the creation of Conservation Units, adding a further 20 million hectares of protected land to the more than 80 million already in existence (totaling 273 units); (d) certification of 87 Native Land areas totaling approximately 18 million hectares; (e) restrictions on to agricultural products emanating from municipalities with the highest rates of deforestation. 2 Economic Incentive policies that use economic mechanisms (prices or otherwise) to incentivize economic agents to reduce deforestation are as follows: (a) the federal police operation Arco Verde in 2008, and (b) special lines of credit in the Northern (FNO, Fundo Constitucional de Financiamento Norte), Northeastern (FNE, Fundo Constitucional de Financiamento do Nordeste) and Central-Western (FCO, Fundo Constitucional de Financiamento Centro Oeste) regions of Brazil for the recovery of degraded areas, reforestation, management, and regularization in the Legal Amazon. ⁎

https://doi.org/10.1016/j.landusepol.2019.104313 Received 21 November 2017; Received in revised form 15 August 2019; Accepted 14 October 2019 0264-8377/ © 2019 Elsevier Ltd. All rights reserved.

Please cite this article as: Bastiaan Philip Reydon, Vitor Bukvar Fernandes and Tiago Santos Telles, Land Use Policy, https://doi.org/10.1016/j.landusepol.2019.104313

Land Use Policy xxx (xxxx) xxxx

B.P. Reydon, et al.

actively moving towards a system of effective governance over land ownership, and in particular, State lands. Although improvements in land governance will not be enough to solve deep, historically rooted issues such as the asymmetries of power between landed and landless groups, or immediately revert one of the most concentrated land distributions in the world, we still argue that establishing effective land governance is a condicio sine qua non for incrementally improving controls over deforestation, preventing land grabs of public land, diminishing land speculation, and struggling against land inequality in dynamic ways such as effective taxation. The paper begins by presenting a brief description of the main causes of deforestation that are identified in the literature on this topic. Next, additional data on deforestation is presented, highlighting that most deforestation occurs on land that is directly or indirectly controlled by the State. The fourth section demonstrates that two problems often relegated to the margins in the literature are, in fact, important drivers of deforestation: land speculation and the absence of land governance. Following on that, the fifth and sixth sections present, respectively, the historical origin and the actual institutional framework of land regulation in Brazil. Section seven brings forth the argument of land governance as an important tool for reducing and controlling the deforestation and it is followed by a description of recent efforts by the government to increase land governance through two main instruments: the ongoing multipurpose cadaster and the Terra Legal regularization program. The eight section briefly updates the recent anti-policies and erosion of the institutional framework and its impacts on the increasing deforestation. Finally, the last section presents summarizes the contributions and discusses the main implementation mechanisms and benefits associated with these proposed changes.

Fig. 1. Annual rates and accumulated deforestation in the Legal Amazon. Source: Brazilian National Institute for Space Research (Instituto Nacional de Pesquisas Espaciais – INPE).

for Amazonian agricultural commodities continues to grow (Garrett et al., 2012), suggesting that Amazon deforestation will remain a persistent problem in the future (Arima et al., 2014). Laurance et al. (2002) were already suggesting that policy initiatives designed to increase immigration and dramatically expand highway and infrastructure networks in the Brazilian Amazon acted as important drivers of deforestation activity. Deforestation is greatest in areas which are most accessible to major population centers and where large-scale cattle ranching and slash-and burn farming are most easily implemented. According to Azevedo-Ramos and Moutinho (2018), the fact that massive sections of the Brazilian Amazon forests are not under effective supervision by a designated public agency (increasing the risk of continued land grabbing), provides further evidence of the absence of effective land governance. Between 2010 and 2015, these undesignated public lands experienced 25% of total accumulated deforestation in Amazon forestlands. Another set of factors underlying deforestation are macro policies and the regulatory turnarounds provoked by government changes in Brazil. After 2014, the federal government’s macro policies became more austerity-oriented. In 2016, major turmoil occurred around the impeachment or coup (depending on one’s point of view), resulting in institutional instability and the deposition of Dilma Rousseff. From 2016 onward, austerity-focused macro policies became more intense and, in relation to deforestation, were also coupled with a conservative push against social and environmental policies. This nexus resulted in weakened institutional capacity, especially as a result of cuts in environmental, social, and science related governmental activities (Petherick, 2017; Magalhães, 2017) The question that remains is: how do we significantly reduce rates of deforestation and put systems in place to make these reductions sustainable over the long-term. Deforestation of the Amazon rainforest is a complex process with multiple drivers, and has been the object of various theoretical and empirical studies.4 Margulis (2001) states: “We do not believe that there is a single, principal force which drives or explains the deforestations in Amazonia. The causes are manifold and result from a sophisticated combination of diverse variables and factors”. The main variables driving deforestation are: (a) benefits associated with the use of land in Amazonia, determined by agricultural prices, increases in the price of land, variation in the prices of inputs, increases in the price of timber, and the reduction in rural wages; (b) public policies and credit – the availability of cheap credit (FINAM – Amazon Investment Fund, FNO – Constitutional Financing Fund for the Northern Region) and fiscal incentives offered by the Amazônia Development Superintendence - SUDAM); (c) accessibility – the construction of highways and/or other works that facilitate access to

2. Deforestation and its principal causes According to the Food and Agriculture Organization of the United Nations – FAO (2010), Brazil has lost an average of 2.6 million hectares of forest per year over the last 10 years, compared with an annual loss of 2.9 million hectares during the 1990s. For comparison, annual forest losses in Indonesia totaled 500,000 ha between 2000 and 2010 and 1.9 million hectares between 1990 and 2000. In 2004, the Brazilian Federal Government started the Action Plan for Prevention and Control of Legal Amazon3 Deforestation (PPCDAm). This is a measure that promotes public policies designed to keep deforestation at bay through monitoring, control, and inspection actions. From the data on Fig. 1, it is evident that after the start of the PPCDAm program, there was a 75% decrease in annual rates of deforestation on the Legal Amazon, from 27,772 km2 in 2004 to 6624 km2 in 2017. Nevertheless, the success of this reduction should not be overstated–the accumulated deforestation in the Legal Amazon is still very large. Accumulated deforestation totaled 167,602 km2 between 1988 and 1997, 190,803 km2 between 1998 and 2007, and 69,991 km2 between 2008 and 2017. From 2017 to 2018, there was an increase of over 8% in deforestation of the Legal Amazon. Summing up, between 1988 and 2018 approximately 436,255 km2 of tropical forest was deforested in the Legal Amazon. From 2002–2011, 32% of this deforestation, (30.000 km2 amazon forest), is attributed to the expansion of soy production in the region (Richards et al., 2014). These findings show that the expansion of areas dedicated to soy production, and the dynamic effects of this and other agricultural activities on local land markets, may incentivize deforestation, again demonstrating the need for effective governance of Amazon forest lands. It is also important to note that the global demand 3 The Legal Amazon (Amazônia Legal) comprises the full extension of the states of Acre, Amapá, Amazonas, Pará, Rondônia, Roraima and Tocantins and parts of the states of Mato Grosso, Maranhão and Goiás.

4

2

For an in-depth review, see Soares-Filho et al. (2005).

Land Use Policy xxx (xxxx) xxxx

B.P. Reydon, et al.

frontier areas; and (d) macroeconomics – cycles of GDP and population growth.5 After the policy interventions of recent years and the crisis of 2008/ 2009, all of these variables have likely acted to increase deforestation, despite overall reductions in rates of deforestation over this period. On the other hand, deforestation increased even during periods in which these variables were not growing. This suggests that there are other, more profound factors at work, whose relative importance has not been highlighted – although implicit factors including political changes and macro policies (especially when oriented towards austerity) do play a significant role in the dynamics of deforestation.

outcomes (less deforestation) in all regions, and public land seems to be particularly vulnerable to negative forest outcomes in South America. Communal land performs well in our Central American cases but worse in Africa, possibly due to the effects of regional conflict and/or weak governance.” We expect that the same patterns persist in Brazil. Nonetheless, as illustrated in Table 2, the largest shares of deforestation occurred on lands controlled by agrarian reform settlements,6 private properties, and areas with no information, each of them responsible for more than 20% of total deforestation. While rural settlements are small properties (in the Brazilian context), there are large numbers of them in the Legal Amazon region, and controlling and monitoring of deforestation in these areas is not fully implemented due to lack of funding for environmental agencies, high costs of monitoring due to remoteness of locations and limited infrastructure, and difficulty in identifying the owners of plots found to exhibit deforestation levels above legal limits. Another factor that may explain the high rates of deforestation on rural settlement land is declining levels of funding for microcredit and other governmental technical assistance for land reform settlers, which could help increase land productivity and enable other land uses that require less deforestation than pasture for extensive cattle ranching. Areas for which no information exists may encompass public or private land, and experienced 22% of total deforestation in 2013and 24,8% in 2015. This category is historically one of the most susceptible to land grabs since neither the state nor private parties are legally owners. Considering private properties, even though deforestation dropped in absolute numbers, it still accounts for over 21% of total deforestation in 2015. The categories of property that should exhibit near zero deforestation due to their status as protected areas (environmental protection areas, Conservation Units, Indigenous Lands), still exhibit small areas of deforestation of between 62 and 245 km2 in each category. Except for on indigenous lands, deforestation increased on all of these land types between 2013 and 2015, with the greatest increases occurring on Federal conservation units (53.3% growth). These data highlight the importance of greater control over land use and occupation in Amazonia. Excluding private property and considering only lands over which the State is custodian (including all types of protected areas, indigenous land, public land, and land reform rural settlements - since the ownership of settlement land is governmental and settlers only have use rights), 78.8% of deforestation in 2015 occurred on land controlled by the State. Evidently, effective control over the different categories of public land would be a decisive first step in the reduction of deforestation in Brazil.

3. Deforestation in detail: by state and type of land access Before discussing the causes behind deforestation and the main steps required to combat it, it is necessary to delve deeper into the characteristics of this phenomenon. In other words, it is necessary to ascertain in which states it is occurring most intensely and if there have been any significant changes in incidence over recent periods. Table 1 presents data on deforestation for each Brazilian state in the Legal Amazon. First of all, Table 1 indicates that there has been a general decrease (of about 4.02% per year) in annual rates of deforestation between 1988 and 2018. This downward tendency holds for 7 of the 9 states in the Legal Amazon: Tocantins (-10.58% per year), Mato Grosso (-6.21% per year), Maranhão (-4.82% per year); Rondônia (-3.44% per year), Pará (-2.86% per year), Acre (-2.67% per year), and Roraima (-1.90% per year). However, there is substantial variability between states. For example, deforestation among the largest deforesters was as high as: 145,539 km2 in Pará, 142,747 km2 in Mato Grosso, and 59,113 km2In Rondônia. Deforestation was least in the states of Amapá, with 1566 km2, Roraima, with 7495 km2 and Tocantins, with 8.648 km2. Another highlight of Table 1 is that, after implementation of the PPCDAm policy in 2004 and 2005, there is a notable decrease of deforestation rates of around 32% for the Legal Amazon states as a whole. The magnitude of this decrease varies by state: Roraima (-57.23%), Mato Grosso (-39.523%), Amazonas (-37.09%), Pará (-33.49%), Amapá (-28.26%) and Acre (-18.66%). The states of Tocantins and Maranhão, on the other hand, experienced increased amounts of deforestation over this period (71.51% and 22.11%, respectively). Fig. 2 illustrates the geographical dimension of deforestation and highlights its association with public land use categories. The figure also shows the large amount of conservation units and indigenous people’s lands in the Legal Amazon, which constitute state land that is largely protected from deforestation. Information on the type of property (tenure category) where deforestation took place is the most important for the objectives of this study. In this regard, Robinson et al. (2014) conducted a complete review of the international literature in an attempt to verify how security of land tenure impacts deforestation. However, the study did draw only general conclusions, stating that “at an aggregate level, the form of land tenure seems to matter in different ways in different regions of the world. We cannot rule out selection or publication bias given our relatively small sample of case studies in each region, but these outcomes emphasize the importance of local factors”. Nevertheless, this study later states that: “overall, protected land is associated with positive

4. Deforestation and land speculation The deforestation of Amazonia, in our interpretation, is the result of the continuation of traditional patterns of expansion of the agricultural frontier in Brazil, which include the following processes: occupation of forested land without clear legal rights or protection (private or public), 6 The issue of Brazilian Amazon deforestation in the agrarian reform settlements is historically controversial. According to Ondetti (2016), there no consensus in the literature on this question, since a several studies have endorsed that the extent of deforestation tends to be greater in agrarian reform settlements than in the surrounding area, whether that be the Amazon as a whole or some sub-region. However, a number of other studies find when forest protection areas established by the state are removed from the comparison, the rate of deforestation is actually somewhat lower within settlements than outside of them - especially because most of expropriated land for agrarian reform programs in the Amazon were already degraded or deforested areas. The extent of deforestation in settlement areas depends on preexisting land use patterns. In some settings, especially those dominated by medium and large-scale cattle and soybean operations, agrarian reform can actually slow forest destruction (Pacheco, 2009).

5 Moutinho et al. (2016) proposed a different group of deforestation causes, which is nonetheless conceptually similar: (a) Brazil’s Growth Acceleration Plan (PAC); (b) growing demands for commodities (beef and grains); (c) unsustainable rural settlement policies; (d) less-than-complete implementation of the Brazilian Forest Code; (e) agribusiness lobby in the National Congress; (f) lack of land tenure and the existence of undesignated public forestlands. It is important to stress that he includes the undesignated public land as an important cause of deforestation.

3

Land Use Policy xxx (xxxx) xxxx

B.P. Reydon, et al.

Table 1 Deforestation for each Legal Amazon state between 1988 and 2017 (in km2). Source: elaborated by the authors with data from Brazil's National Institute for Space Research (Brazil's National Institute for Space Research (Instituto Nacional de Pesquisas Espaciais – INPE). Year

AC

AM

AP

MA

MT

PA

RO

RR

TO

Total

1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 i1

620 540 550 380 400 482 482 1208 433 358 536 441 547 419 883 1078 728 592 398 184 254 167 259 280 305 221 309 264 372 257 444 −2.67***

1510 1180 520 980 799 370 370 2114 1023 589 670 720 612 634 885 1558 1232 775 788 610 604 405 595 502 523 583 500 712 1129 1001 1045 −0.66NS

60 130 250 410 36 0 0 9 0 18 30 0 0 7 0 25 46 33 30 39 100 70 53 66 27 23 31 25 17 24 24 0.37NS

2450 1420 1100 670 1135 372 372 1745 1061 409 1012 1230 1065 958 1085 993 755 922 674 631 1271 828 712 396 269 403 257 209 258 265 253 −4.82***

5140 5960 4020 2840 4674 6220 6220 10391 6543 5271 6466 6963 6369 7703 7892 10405 11814 7145 4333 2678 3258 1049 871 1120 757 1139 1075 1601 1489 1561 1490 −6.21***

6990 5750 4890 3780 3787 4284 4284 7845 6135 4139 5829 5111 6671 5237 7510 7145 8870 5899 5659 5526 5607 4281 3770 3008 1741 2346 1887 2153 2992 2433 2744 −2.86***

2340 1430 1670 1110 2265 2595 2595 4730 2432 1986 2041 2358 2465 2673 3099 3597 3858 3244 2049 1611 1136 482 435 865 773 932 684 1030 1376 1243 1316 −3.44***

290 630 150 420 281 240 240 220 214 184 223 220 253 345 84 439 311 133 231 309 574 121 256 141 124 170 219 156 202 132 195 −1.90**

1650 730 580 440 409 333 333 797 320 273 576 216 244 189 212 156 158 271 124 63 107 61 49 40 52 74 50 57 58 31 25 −10.58***

21050 17770 13730 11030 13786 14896 14896 29059 18161 13227 17383 17259 18226 18165 21650 25396 27772 19014 14286 11651 12911 7464 7000 6418 4571 5891 5012 6207 7893 6947 7536 −3.93***

Notes: AC: Acre. AM: Amazônia. AP: Amapá. MT: Mato Grosso. MA: Maranhão. PA: Pará. RO: Rondônia. RR: Roraima. TO: Tocantins. 1Average rate of annual growth in percentage. It is as estimate of the coefficient of a log-linear regression against time. In this case, the t test reveals the existence or not of a tendency in the data. *** significant at 1%, ** significant at 5%, NS indicates that it is not significant.

extraction of timber, introduction of livestock farming,7 and finally, the development of more modern forms of agriculture. These economic activities generate income and legitimize land occupation by new owners in the short term without requiring nearly any prior resources.8 In the long term, affected lands either remain under intensive livestock farming or, if the demand exists, are converted to grain production or other economic activities. The most important factor driving this process is the expectation that there will be future demand for this land,9 which in turn inflates its present price. The closer the land is located to productive regions, the higher the price (Telles et al., 2018). Appreciation in land values thus accelerates as expectations of future profitability grow. While the question of land speculation appears in studies conducted by Margulis (2001, 2003), as well as in the literature already cited, it is usually associated with increases in land price. Nevertheless, increases in land price do not necessarily guarantee that a speculative process is underway (Telles et al., 2016). In fact, the average price of land in Brazil’s Northern region approximately tracks movements in land prices

in the rest of the country, and does not generate substantial speculative gains. Land speculation which, as this article proposes, is the driving force behind deforestation of the Amazon rainforest, occurs at a far more microeconomic level and is associated with the actual occupation of land, rather than mere speculative investments. This can be seen very clearly in the existing field research. In practice, anybody who acquires or occupies land that includes forest has a clear understanding that his land, their investment, will grow in value in proportion to the deforestation process. As shown in Table 3,10 the price of forested land ranges, in the different states, from R$ 108 per hectare in Acre to R$ 546 per hectare in Mato Grosso. It is evident from this table that the least deforested states have the lowest land prices, while the states of Mato Grosso, Pará and Rondônia have the highest prices. The most important conclusion to be drawn from Table 3 is that, across all states in the Brazilian Amazon, deforestation consistently raises property values. On average, deforestation increased the value of land more than fourfold. This happens because the price of land derives from expectations of future productive gains arising from associated agricultural activities on this land. Deforested areas can be used immediately without incurring costs associated with clearing the forest.

7

Sparovek et al. (2019) shows that the main driver of conversion of forest to agriculture is the existence of large amounts of vacant, potentially productive land that can be appropriated at low cost. The convergence of these factors turns deforestation into an unbeatable capital appreciation strategy. 8 It is these occupiers that frequently make use of slave labor. 9 This arises as a result of an increase in the price of beef cattle or soy, or even the announcement that Brazil may become the largest producer of alcohol in the world. Recently, these expectations have converged, making demand for land, and its price, grow even more, and consequently creating ever more pressure to participate in deforestation.

10 The methodology utilized by Informa Economics involves collecting average prices in homogeneous regions in the cited states using a non-uniform terminology. To forest we sum areas of so-called forest, forests that are easy to access and those that are difficult to access. For pasture land we use formed pasture (easy and difficult access), formed high-maintenance pasture, and formed low-maintenance pasture.

4

Land Use Policy xxx (xxxx) xxxx

B.P. Reydon, et al.

Fig. 2. Deforestation and main public land use categories in the Brazilian Amazon. Source: Moutinho et al. (2016). Table 2 Deforested area in the Amazon by land title category from 2012 to 2015. Source: Moutinho et al. (2016). Agrarian Category

Indigenous Lands Federal Conservation Unit State Conservation Unit Environmental Protection Areas Rural Settlements Private properties Federal Public Lands State Public Lands No information Grand total

2012

2013

2014

2

2

2

2015

(Km )

(%)

(Km )

(%)

(Km )

(%)

(Km2)

(%)

168 175 117 124 1239 986 574 15 982 4381

3,8 4 2,7 2,8 28,3 22,5 13,1 0,3 22,4 100.0

170 187 175 228 1518 1,009 743 31 1222 5282

3,2 3,5 3,3 4,3 28,7 0 14,1 0,6 23,1 100.0

71 120 174 202 1269 883 584 0 1047 5350

1,6 2,8 4 4,6 29,2 20,3 13,4 0 24,1 100.0

62 184 233 245 1437 1113 670 7 1306 5256

1,2 3,5 4,4 4,7 27,3 21,2 12,7 0,1 24,8 100.0

In the most extreme case, in state of Acre, deforestation multiplies land by more than 14 times, while in the state of Amazonas, the multiple is almost 10 times. Very few alternative investments yield such high returns. It should be borne in mind that land owners participating in deforestation benefit not only from these increases in the value of their land, but also make gains from the sale of timber (in Cotriguaçú, in state of Mato Grosso, net timber returns are estimated at around R$ 2400 per hectare) and also from the land’s subsequent economic use (if this is due to livestock farming, it will generate additional net revenue of over R$ 120 per hectare per year11). Therefore the primary catalyst of

Table 3 Average prices of forest and pasture land in Brazilian Legal Amazon states (R$ per hectare, in December 2015). Source: Informa Economics. States

Forest (R$/ha)

Pasture (R$/ha)

Increase (%)

Acre Amapá Amazonas Pará Rondônia Roraima Mato Grosso Average Northern Region of Brazil

534,25 470,00 574,33 1.715,09 2.453,40 817,00 3.005,8 1.367,12

3.246,80 1.006,50 1.448,38 2.976,84 7.288,89 1.277,50 6.906,2 3.450,16

608 214 252 174 297 156 230 252

11

5

See Margulis (2003).

Land Use Policy xxx (xxxx) xxxx

B.P. Reydon, et al.

Fig. 3. Legal status of lands in the Amazon, from data in the National Rural Cadaster System– SNCR and Protected Areas, 2007. Source: Barreto et al., 2008.

deforestation is the combination of gains from the appreciation of land values, the conversion of the land from forest to productive uses, and gains from timber and livestock farming. This process of acquisition and deforestation, which is already extremely profitable in private areas, is becoming much more lucrative on vacant lands which, according to estimates,12 represent 42% of the total area of Amazonia. This is where the majority of deforestation currently takes place. On vacant lands, the economic gains from timber, livestock farming, and the appreciation of land values are multiplied, since the land itself was simply usurped from the public domain, rather than purchased.13 Thus, only with better land governance, including a regular land cadaster, organized registration procedures, and land taxes,14 will it be possible to diminish land speculation.

registration system is also questionable. Of the 178 million hectares declared as private property, 100 million may be based on fraudulent documentation. A further 42 million hectares of this category are declared as “possessions” within the registry. These possessions may or may not be feasible for regularization, depending on their unique circumstances, including size, history, and location.15 As a result of these challenges to regularization, 30% of land area is under registration that is legally uncertain and/or contested. The Brazilian State is so aware of its inability to regulate the use of land, including the dysfunctionality of its land cadaster, that it was obliged to take the following concrete actions to reduce deforestation in Amazonia and improve land governance. These measures were introduced on an emergency basis without addressing the problem at its source: (a) the introduction of Law 11952/09 regularizing land possessions of up to 400 ha at no cost and selling possessions of between 401 and 1500 ha (with squatters required to demonstrate that they have lived on the land since 2004); (b) creation of numerous protected areas (APA) in the form of Conservation Units (based on Law 9985 of July 2000), which were focused on protection along the margins of major main highways under construction in the Amazon region, (c) creation of the Rural Environment Registration (CAR),16 under the auspices of the Brazilian Forest Code, to oblige owners to georeference their properties, with the goal of being able to identify properties, their owners, and their respective Legal Reserves. Recently, Sparovek et al. (2019) integrated Brazilian official land use tenure data into a single map, including all the data from the CAR, treating the overlaps with a carefully constructed hierarchical classification. Their results show that 36% of lands are public, 44% are private, and 17% are unregistered or with unknown tenure. Another important issue pointed out is the fragilities in Brazilian land tenure evidenced by the extent of overlaps among land tenure categories (50% of the registered territory) and that one sixth of Brazil (16.5%) is not classified as having any official land tenure registration – areas that usually spark

5. Land appropriation in Amazonia and the insecurity of property rights The practice of land appropriation is associated with the absence of effective regulation of land property in Brazil, and particularly in Amazonia. Existing registration data, based on the declarations of landowners registered with the Brazilian National Institute for Colonization and Agrarian Reform (Instituto Nacional de Colonização e Reforma Agrária – INCRA), show that, in 2003, 35% of the 509 million hectares of land in the Legal Amazon were occupied under the right of private possession, either as registered property or as possession. On the other hand, the recent process of creating Federal or State reserves of different types means that today, 42% of the Legal Amazon is under some form of protection. Approximately half of this protected area was Native Land, while the other half consisted of Conservation Units of various types. The remaining 24% of Amazonian land did not belong to any of these categories and is technically therefore considered to be unallocated public land (Fig. 3). On the ground, the situation is more complex and uncertain than these numbers suggest. Much of the land in the Protected Areas is physically occupied by private users, whose claims to occupation may or may not be valid, depending upon the complex legislation that governs these topics. The large area described as “private” by the

15 The INCRA cadaster includes all types and sizes of squatters, both small holders with less than 200 ha and those with over 1,000 ha. 16 The enrollment of all rural land in the CAR is mandatory and it represents the first step towards environmental regularization. It is a public cadaster at a national level and aims to integrate environmental data from the parcels (permanent protection areas (APP), restricted areas, Legal Reserve areas (Reserva Legal), native vegetation indicators, and others. It Works as a database for control, monitoring and environmental and economic planning tool, especially important as a tool to control deforestation. The main drawback of the CAR structure is that information provided by the owners (or supposed owners) is declaratory and keeping it up to date is still a problem.

12 Estimates by Shiki and Shiki (2011) show that 42% of land in Amazonia is considered vacant land in their typology. 13 Perhaps some expenses are incurred, including the hiring of thugs and weapons, and the legal and illegal costs of regularizing the area. 14 See Reydon et al. (2015) and Reydon and Fernandes (2017) for more information on the issue of land taxes in Brazil.

6

Land Use Policy xxx (xxxx) xxxx

B.P. Reydon, et al.

disputes that result in deforestation and violence. In this context, the CAR is an important indicator to assess the land tenure status in the Amazon region, as a large part of rural properties do not have a proper land title (Homma, 2010). Also, the CAR is necessary to access most of the credit lines for agriculture. From the legal point of view, the links between lack of control over landownership and deforestation can be seen in the report produced by the Parliamentary Inquiry Committee on Land Grabbing (CPI da Grilagem de Terras of 2001). That report, as well as a study by INCRA (1999), showed that, in practice, it was very easy to obtain titles and to register land in the Amazon region. After the publication of the CPI report, a number of actions were taken to reduce land grabbing. According to Lima (2002), the equivalent of 48.5 million hectares in fourteen districts of the state of Amazonas had their registration cancelled.17 The CNJ (Conselho Nacional de Justiça) canceled a further 410 million hectares of rural property registrations in the state of Pará in 2009 (totaling more than three times the territory of the state). Two conclusions may be drawn from these initiatives: first, that there are sectors within relevant government agencies that proved themselves capable of developing innovative ways to curb illegality and land grabs, which is notable and should be reinforced; second, that the State must develop and implement effective policies to ensure reasonable governance over its own lands and, through these policies, diminish illegal land appropriation, fraud, and other kinds of land grabbing. Although the Brazilian government has made some slow progress towards the enforcement of the rule of law over its own lands, limitations in its knowledge and effective control over landownership and land use is still a major problem, especially considering the resulting impacts of land use on deforestation in the Amazon region.

register, which, ideally, would be able to define and register vacant lands and permit their use through agrarian policies. Prior to the 1850 Land Law, the registration of property was done with the Parish Land Registries, under the responsibility of the local vicar (this process, in turn, began with the Law of 1822, upon the demise of the sesmarias). Parish registration continued to be used for a long time after the proclamation of the Land Law. In 1864, a land law decree obliged all holders of land to register their possessions in the vicar’s registry, though this was never enforced. The change from monarchy to a federative republic in 1889 delivered the new rights to the states over the use and management of vacant lands, generating the possibility that state representatives might transfer these lands by granting unregistered titles. This happened to a greater or lesser degree in different states, but in all cases created yet another ambiguity over the granting of titles.20 Furthermore, even if the de jure regulations called for every state to create a Land Institute in order to manage, map, and register their public vacant lands (terras devolutas), this only occurred in practice on an ad hoc basis. The fact that those vacant lands were seldom brought under state management is explained through the fact that it was (and remains) of interest to involved parties to allow those lands to remain free for private appropriation. Considering another perspective, the political power of private interests was always capable of applying pressure on regulators to leave public lands unregistered, in order to allow the later private appropriation of these lands without cost. Functional Land Institutes do exist in some of the Amazonian states, but this is far from being the rule. There is a great lack of information about their operations and there is too little transparency – not to mention that there are no qualified academic studies on the theme, severely hindering scientific exploration of their actions. Table 4 displays the names of the state Land Institutes and the law of their creation, providing support for the argument that they were created (or dissolved) on an ad hoc basis. The institutionalization of the Public Land Registry in 1900 was, arguably, the main step toward the creation of the property registration system that prevails today. This ruling stated that all land users must demarcate and register their rural or urban properties, though it did not establish any formal audit procedures, nor did it create a formal registry. Departing from the 1900 act, the mainstream interpretation was that the State would also need to demarcate and register its (vacant) lands, which proved impractical given that these lands are defined by process of elimination. Consequently, the State itself is acting illegally by failing to comprehensively register its own lands. This lack of registration has the effect of increasing the possibility of fraud in land registration at public registry offices. The proclamation of the Civil Code of 1916 reasserted the registry office as the institution of registration and allowed public lands to be the object of adverse possession. In the words of Silva (1996): “with this, the framework for the transformation of the State into an owner like anyone else, was complete. And thus, the doctrine of statute of limitation over vacant lands was sustained. In other words, the possibility of adverse possession of vacant lands”. The Brazilian Civil Code, therefore, through motives not necessarily linked to the interests of landowners, introduced important milestones in the institutionalization of land access in Brazil. In particular, it required the registration of property in official registries, which was in some cases sufficient to prove ownership. In a way, registration at the

6. The institutional framework of land regulation: why is land governance weak? This section analyzes the historical creation of Brazil’s institutional land governance framework, which has led to the absence of land ownership and use regulation. This section will demonstrate that many of the characteristics underlying today’s institutional shortcomings are inherited from the Brazilian State’s historical ineptitude regarding land governance. Before the Brazilian Land Law of 1850, the rules regarding the occupation of urban and rural lands were defined based upon the powers of the king, the Church or the political and physical power of the occupants. The 1850 Land Law should be understood in a more general context of laws that restricted access to land in colonial possessions.18 In keeping with the interests of landowners in the country, the Land Law made it possible to regularize possession,19 the fruit of the occupation of vacant lands, which rendered unviable the creation of a regular cadaster. Put more directly, the negative definition of public vacant lands (terras devolutas) as being all land that is not privately occupied, and the fact that in a given moment of time it was not possible to map or register all private lands, the registration and mapping of terras devolutas was practically impossible, as it could always be contested by a private occupant. The legal possibility of regularizing the possession of vacant lands has existed for centuries in Brazil. In addition to adverse possession (which establishes that the squatter may regularize their property after a number of years), the states themselves (primarily after the creation of the Republic) occasionally granted property with or without title. These informal or semiformal grants are one of the main historical factors that have made it near impossible to establish an effective land

20 There was also a failed attempt at regulating property via the Torrens Registry (1891), in which squatters and owners would be able to obtain definitive title through an uncontested petition. On the other hand, the possibility of legalization of squatter’s possession in 1895 and in 1922 (in regard to possession between 1895 and 1921) had the effect of creating the conditions for squatter’s possession to persist and for land market regulation, as expressed in the Land Law of 1850, to be weakened.

17 Reydon et al. (2015) provides substantial evidence that land tenure or land governance in the country, particularly in the Amazon, is chronically weak. 18 As in Latin America, Australia and the USA. 19 i.e. transformation of possession into property through legal channels.

7

Land Use Policy xxx (xxxx) xxxx

B.P. Reydon, et al.

Table 4 Brazilian Legal Amazon states and their Land Institutes. State

Land Institute

Law and date

Acre Amapá Amazonas Maranhão Mato Grosso Pará Rondônia Roraima Tocantins

Land Institute of Acre State (Instituto de Terras do Acre – ITERACRE) Land Institute of Amapá State (Instituto de Terras do Amapá – ITERRAP) Land Policy Bureau of Amazon State (Secretaria Estadual de Política Fundiária – SPF) Colonization and Land Institute of Maranhão Sate (Instituto de Colonização e Terras do Maranhão – ITERMA) Land Institute of Mato Grosso State (Instituto de Terras do Mato Grosso – INTERMAT) Land Institute of Pará State (Instituto de Terras do Pará – ITERPA) Land and Colonization Institute of Rondônia State (Instituto de Terras e Colonização de Rondônia – ITERON) Land and Colonization Institute of Roraima (Instituto de Terras e Colonização de Roraima – ITERAIMA) Agrarian Development and Land Regularization Bureau of Tocantins State Secretaria de Desenvolvimento Agrário e Regularização Fundiária (SEDARF)

1,373/2011 1,184/2008 4,163/20151 6,272/1995 3,681/1975 4,584/1975 203/1988 30/1992 2,730/20132

Notes: 1 The Instituto de Terras do Amazonas (ITEAM) was dissolved in 2015. 2 The Instituto de Terras do Estado do Tocantins (ITERTINS), created in 1989, was dissolved in 2013.

registry office gave an appearance of legality to the property without the existence of any mechanism to guarantee it.21 The major institutional innovation in the area of agrarian policy and administration in Brazil was the Land Statute (Estatuto da Terra) of 1964, which established rules and concepts that continue to prevail to the present day. To guide the implementation of agrarian and agricultural policy, the Statute of 1964 created the Rural Property Registration.22 Under this system, all private or public property should be registered, including land possessions. The owners should provide information on the situation of documentation and use of land (used to estimate productivity) in order to facilitate agrarian reform. INCRA, created in 1970, became responsible for the management of the National Rural Cadaster (SNCR), which maintained the Rural Property Registration. Once the property was registered, INCRA would issue a Rural Property Registration Certificate (CCIR), which in turn was required for any type of land transaction. Squatters registered by INCRA also received the CCIR and would have to pay the Rural Property Tax, though the value of these taxes was always kept at a low level. The Land Statute once again maintained legitimacy of possession, thereby permitting entitlement to informally occupied public lands. Fig. 4 aims to summarize, from a schematic point of view, the interrelationships between agencies in the Agrarian Administration system in Brazil. From this table, one can see that there are no links between INCRA and municipal governments, creating many problems and confusion along the intersections between rural and urban lands. Moreover, there is no single institution that centralizes land registrations and provides a link with the Judiciary bodies responsible for property entitlement. Furthermore, there are many forms of vertical and horizontal overlaps between agencies due to the lack of articulation and communication between them. The system is further complicated by the lack of a unified land cadaster, resulting in extended bureaucratic processes and the weakening of land claims and land rights

(Paixao et al., 2015). A large proportion of both rural and urban agrarian problems in Brazil that are not resolved in the administrative domain end up in court. Due to case overloads and other complexities, these cases often take years to process, meaning that land-related cases, whether rural or urban, are often tried as faits accomplis. Therefore, the major problem of deforestation in Amazonia is associated with Brazil’s weak system of land governance, which is itself the result of an historical process that generated inadequate legal and institutional frameworks. Only with the construction of an institution whose goal it is to improve land governance and build an appropriate legal framework in Brazil will it be possible to reduce deforestation and have an adequate use of land in the country. 7. The need for land governance23 as a prerequisite for the reduction of deforestation Agriculture in Brazil is exemplary, with growth in food production, supply of energy, and foreign currency earnings, and greater international insertion in recent years. Nevertheless, conflicts and irregularities associated with land ownership and land grabs of public lands remain a central problem, particularly in Amazonia. Our argument, as stated in the introduction, is that land governance is a necessary condition for reducing or controlling deforestation. Before proceeding, it is necessary to clarify what the authors understand as “land governance”. Without plunging too deep into the academic discussion over different meanings and definitions, we understand land governance as concerning “the rules, processes and structures through which decisions are made about the use and control over land, the manner in which the decisions are implemented and enforced, and the way that competing interests in land are managed” (Palmer et al., 2009). This definition also encompasses statutory, customary, and religious institutions and formal and informal actors. It is, fundamentally, a viewpoint about power and the political economy of land, since “the power structure of society is reflected in the rules of land tenure” and “the quality of governance can affect the distribution of power in society” (op. cit.). Beyond this definition’s consideration of the institutional framework, it also encompasses the social legitimacy and power asymmetries present in relation to the land.24

21 The most common irregularities are the granting of titles for nonexistent or vacant properties and the superimposition of various areas, i.e. various owners having tile over the same land. When this occurs, it is said that the land has ‘floors’: for every owner with an irregular title in that area, an extra floor is added. The federal government is taking a decisive step in the regulation of the rural and urban land market by implementing, not without some difficulty, Law 10267/2001 in which the registry offices are obliged, whenever there is a change in property, to transfer it to INCRA on a plan with its boundaries in map form (latitude and longitude). 22 As the 1967 registration and subsequent re-registrations were for fiscal purposes (via rural land tax - ITR), were based on the declarations of the landowners, and were not audited, this is not very reliable, as documented in the study by Wilkinson et al. (2012). Other recent attempts to integrate the cadasters of the various public agencies in order to improve the quality of information have failed on account of the absence of political will and of an agency that is prepared to take on the role of carrying out agrarian governance in the country.

23

“Governance” refers to the system of values, policies, and institutions by which a society manages its economic, political, and social affairs through interactions within and between the state, civil society, and the private sector. Land governance concerns the rules, processes, and organizations through which decisions are made about access to land and its use, the manner in which the decisions are implemented, and the way that competing interests in land are managed. 24 Other definitions that are reasonably in line with the one presented can be found in: Deininger and Feder (2009); Borras and Franco (2010); Deininger (2011). 8

Land Use Policy xxx (xxxx) xxxx

B.P. Reydon, et al.

Fig. 4. Institutions responsible for Land Administration System in Brazil. Source: Reydon et al (2015).

The benefits from an adequate system of land management depend on clear identification of registered properties and a simple, effective mechanism to obtain registration information and keep it up-to-date. Only with the effective governance of land, particularly the creation of a modern, self-perpetuating registry, will it be possible to: (a) guarantee the rights of private property for different ends, including business, leasing, and credit guarantees for the granting of payments for environmental services, amongst others; (b) identify public land and guarantee its adequate use for the creation of reserves, settlements, or colonization; (c) establish effective agrarian policies, including agrarian reform, agrarian credit, and taxation of land; (d) regulate land purchases to limit access for foreign stakeholders or owners who already have substantial land holdings,; (e) zone the use of land, that is, establish and regulate land zoning of agricultural and livestock production in certain regions, and establish protected areas and prohibition of deforestation; (f) regulate the processes of conversion of agricultural land into urban land and therefore establish a register for the collection of taxes on property (IPTU and ITR). Land governance alone will not solve the problem of deforestation in Amazonia, but it is a prerequisite25 for addressing the problem. In regard to vacant lands, formal registration would allow the State to

identify and control them and prevent inappropriate private appropriation and deforestation. Regularization of vacant lands would also make it possible to use these lands in the execution of agrarian policy in Brazil, including organized colonization, agrarian reform, and other uses. On private lands, effective, participatory governance based on accurate local knowledge will allow for establishment of priorities for use and adequate enforcement, planning, and regulation of soil use. Moreover, through zoning and other compulsory tools, registration of vacant lands will prevent deforestation and will certainly limit land speculation, which is the main cause of deforestation. 8. Recent advances in land governance in the Amazon Region Incentivized by the FAO’s Voluntary Guidelines, land administration and land governance of both public and private lands in Brazil is improving along a number of dimensions. Major advances include a) improvements in the land cadaster26 resulting from law 10.267/2001; and b) regularization of both rural and urban land ownership. The following analysis only considers the impacts of these changes within the Amazon region. 8.1. Land cadaster in Brazil: CNCR, SIGEF

25

Taking the payment for environmental services and land governance as a pre-requisite as an example, Robinson et al. (2014) in a seminal article about land tenure and forest conservation, argues that: “However, in light of PES programs and REDD, where future incentives are tied to particular land usebased outcomes (e.g. maintain forest), the security of tenure is crucial to influence landholders’ decision-making. (…). Therefore, security is necessary to prevent deforestation through market-based conservation mechanisms, but alone does not necessarily protect forests.”

The enactment of Law 10,267/2001 created the georeferenced CNIR (National Rural Property Cadaster). The law requires land owners to 26 The CAR and SICAR systems are also part of this, but will not be analyzed here due to the impossibility of comparing them with the cadastral base from SIGEF, at least at the time this article was written.

9

Land Use Policy xxx (xxxx) xxxx

B.P. Reydon, et al.

Fig. 5. Public and private parcels georeferenced in the CNIR Cadaster in 2015. Source: Brazil’s National Institute for Colonization and Agrarian Reform (Instituto Nacional de Colonização e Reforma Agrária – INCRA).

georeference their property when registering any change to it with a notary. This georeferenced information is sent to both INCRA and the Brazilian Federal Revenue Service (Receita Federal do Brasil – RFB), as well as the existing main cadasters (SNCR and CAFIR). As Guedes and Reydon (2012) showed, the system only began to show success in the registration of private properties around 2010, at which point the land ownership mosaic began to emerge. In 2013, an important innovation related to the Terra Legal program was introduced: the Land Management System (Sistema de Gestão Fundiária – SIGEF). SIGEF is an electronic tool developed by INCRA and MDA, which automates various procedures and has since mapped 61% of Brazilian lands. INCRA’s Land Data Collection (Acervo Fundiário) is a visualization tool that allows the general public to access, download, and visualize 3485 million hectares of georeferenced parcels (public and private). Together with the information from other government agencies (FUNAI – indigenous land, ICMBio – federal and other protected areas, for instance) this area encapsulates 522.4 million hectares, or 61.3% of the Brazilian territory, excluding overlaps. Fig. 5 below illustrates the dimensions of certified parcels (public and private) in Brazil, while Table 5 reports summary data from

INCRA’s Land Data Collection related to the georeferenced parcels. These advances represent a significant improvement in Brazilian land governance. Nevertheless, the main unresolved question is the consolidation of the CNIR and the practical integration of all existing cadasters related to land – i.e. the construction of an integrated, comprehensive land cadaster. 8.2. Land destination and the terra legal regularization program Legal Land (Terra Legal), a federal program, was designed to tackle problems related to the large amount of land lacking clear property rights. The program’s main mechanisms include the regularization of both private lands and the destination of land to public uses. Law 11,952, passed in July 2009, provides procedures for the regularization and titling of individual or familiar land holdings on federal public land located in the Legal Amazon territory. First of all, the law requires the Brazilian Agrarian Development Ministry (Ministério do Desenvolvimento Agrário – MDA) to transfer lands without destination to municipalities, provinces, and other federal institutions for the regularization of urban plots, indigenous lands, protected areas, land reform settlements, quilombola settlements, and

10

Land Use Policy xxx (xxxx) xxxx

B.P. Reydon, et al.

These results, though below the aim of the program, demonstrate that titling is a productive way forward. There is a substantial need for more regularization of public lands.27

Table 5 INCRA’s Land Data Collection related to the Brazilian georeferenced parcels in September of 2016. Source: Brazil’s National Institute for Colonization and Agrarian Reform (Instituto Nacional de Colonização e Reforma Agrária – INCRA). Type INCRA own information Settlement projects Quilombola lands Certified public properties Certified private properties Land regularization agreements Subtotal Partner entities databases Indigenous Lands (TIs) Protected Areas (UCs) Georeferenced polygons from SRA/MDA1 Georeferenced polygons from Terra Legal2 Subtotal Grand total Net area without overlaps Brazil surface área

Properties

Area (ha)

7.796 312 9.800 263.038 107.853 388.799

76.907.386 2.323.928 86.554.346 178.342.807 4.335.994 348.464.461

40,92

588 1.481 80.041 148.969 231.079 619.878

116.625.185 152.029.511 3.254.260 23.976.310 295.885.266 644.349.727 522.397.153 851.576.700

34,75 75,67 61,34 100

9. Rise of deforestation under current policy scenario

%

Brazil was considered a global example in terms of policies, agreements and public policies to reduce deforestation, especially in tropical forests of the Amazon region (Gibbs et al., 2015; Gibbs et al., 2016; Rausch and Gibbs, 2016). After 2012 there has been an up-turn in deforestation mainly because of several land governance failures in the implementation of the principal agreements and Brazilian laws that should be keeping clearing rates under control (Carvalho et al., 2019). According Ferrante and Fearnside (2019), Jair Bolsonaro, who took office on 1 January 2019 as Brazil’s new president, has taken actions and made promises that threaten Brazil’s Amazon forest, as the encouraging the illegally deforesting. The newly elected far-right government of Jair Bolsonaro is fulfilling its election promises made to large landholders and mining industry sectors by Slashing the budget of scientific research and environmental protection agencies - with extended impact of freezing or dismantling proven efficient policies against deforestation – and investing against the rights of indigenous lands and other forested protected areas. Bolsonaro’s party and its allied base are pushing forward a series of bills and constitutional amendments in the National Congress that, if approved, would result in an effective threat to the environmental protection system that was painstakingly created over the last decades. This movement is aimed at reducing the restrictions of environmental licenses for new infrastructure projects, mining and other economic activities while reducing the protection of indigenous lands and protected areas, seeking to open these areas to private exploitation in the near future (Reyes-Galindo et al., 2019). The anti-environmental agenda is advancing at large steps and has impacts beyond deforestation – the regulations for use and sale of pesticides were weakened, a new law is in discussion to allow hunting wild animals (Abessa et al., 2019). Beyond that, Bolsonaro government is dismantling institutions related to environmental protection and traditional and indigenous populations land rights through merging or changing ministries attributions (in January 2019 he transferred the Brazilian Forest Service and the creation of indigenous lands to the responsibility of the Ministry of Agriculture) and focused budget slashes (Campos-Silva and Peres, 2019; Begotti and Peres, 2019). Brazilian environmental policy is at stake at the moment this is being written and, although there is a lot of uncertainty, all the recent reports points out it is surely is going downhill.

Notes: 1 Land Reorganization Bureau of Brazilian Agrarian Development Ministry (Secretaria de Reordenamento Agrário do Ministério do Desenvolvimento Agrário). 2 Land regularization program.

other land uses of public interest. For these purposes, the law also created the Legal Amazon Land Regularization Special Office (SERFAL), which falls within the MDA hierarchy. The main objective of Law 11,952 is the regularization and titling of landholders without legal titles (adverse prescription) on lands of up to 15 fiscal modules (the maximum size varies depending on the municipality, but the area limit is 1,650 ha, for a fiscal module of 110 ha). The basic requirements are that the landholder cannot have other land titles and must maintain the land under productive use. Initially, the federal government forecasted the titling of 300,000 rural and urban land occupants over a 3-year period in 463 municipalities located on the Legal Amazon. This would represent 67 million hectares of public land that could be destined to landholders via regularization – equivalent to the area of Germany and Poland together, or 13.42% of the Brazilian Legal Amazon area (Oliveira, 2016). Until now, the main actions developed by Serfal included identifying, georeferencing, and digitalizing federal lands, as prescribed by Law 10,267/2001, which constituted an important contribution since this information was not previously available. Data from the Legal Amazon Land Regularization Special Office (Serfal, 2015) shows that, between 2009 and 2015, a total of 113 million hectares of federal public land in the Legal Amazon were identified, around 51 million were georeferenced and certified, and 29.7 million were registered at the register offices, as can be seen in Table 6. Table 6 also shows that, out of the total land under the Terra Legal program, approximately 58 million hectares (51.3%) has been destined, 21 million (36.2%) of this has been georeferenced, and 11.4 million (19.7%) has been registered by a notary. Table 7 shows that, from the lands yet to be destined, 30 million hectares (54%) have been georeferenced and certified and 29.7 million hectares (33.3%). have been registered. Analyzing the results from 2009 to 2014, the Terra Legal program georeferenced almost 10 million hectares of federal public land in the Legal Amazon, bringing the total area of georeferenced federal public lands to 51 million hectares, or45% of the total federal public land in the Legal Amazon. Furthermore, 11 million hectares of land were titled through regularization during this period, which represented approximately 17% of the initial forecast (56 million hectares). Even with the great increase of titles emitted in 2014, we observe that titling is underperforming, since the total number of titles emitted contrasted with the initial forecast.

10. Final remarks This article began by debating the main causes of deforestation in the Brazilian Amazon rainforest. It presented data demonstrating that states with the greatest increase in deforestation are indeed the expected ones: Pará and Mato Grosso. Another important conclusion is that most deforestation, at least during the years of 2012 and 2013, occurred on lands either directly or indirectly controlled by the State, thus demonstrating that a lack of land governance is a very important part of the deforestation dynamics. Section 4 showed that, besides the macro determinants of deforestation, there are important micro determinants, including land speculation, which in turn is driven by the considerable increase in its value that comes from deforestation. Section 5 and 6 reaffirmed that the deforestation in the Amazon results 27 In the last year some progresses in this field took place, as meetings of the representatives of the State Land Institutes and other ministries related to land issues created the Cartas of Palmas, some principles to be followed by them.

11

Land Use Policy xxx (xxxx) xxxx

B.P. Reydon, et al.

Table 6 Federal public land destination in the Legal Amazon in 2015 (millions of hectares). Source: Legal Amazon Land Regularization Special Section (Subsecretaria Especial de Regularização Fundiária na Amazônia Legal – SERFAL).

Federal public land destined Federal public land yet to be destined Total Legal Amazon total area

Total areas (ha)

Georeferenced and certified areas (ha)

Registered at the land notary (ha)

58 55 113 501.6

21 30 51

11.4 18.3 29.7

References

Table 7 Areas assigned and titles emitted by Terra Legal (TL). Source: Reydon (2019). Titles emitted by Terra Legal (TL) Land assignment Federal lands assigned by TL until January 2016 Federal lands assigned by TL to FUNAI (indigenous land) Federal lands assigned by TL to INCRA Federal lands assigned by TL to SPU Federal lands assigned by TL to MMA (conservation units) Federal lands assigned by TL to MDA (land reform / tenure regularization) Land currently being studied by TL Land yet to be consulted Assigned federal lands before entering into force of CT Land titles emitted Land titles emitted until January 2016 Total Federal lands in the mandate of TL

Million ha

%

38,202,778 2292 92,047 55 6,271,543

67.50% 0.00% 0.20% 0.00% 11.10%

31,836,841

56.20%

8,288,868 6,808,354 2,000,000

14.60% 12.00% 3.50%

1,300,000 56,600,000

2.30% 100%

Abessa, D., Famá, A., Buruaem, L., 2019. The systematic dismantling of Brazilian environmental laws risks losses on all fronts. Nat. Ecol. Evol. 3, 510–511. https://doi. org/10.1038/s41559-019-0855-9. Arima, E.Y., Richards, P., Walker, R., Caldas, M.M., 2011. Statistical confirmation of indirect land use change in the Brazilian Amazon. Environ. Res. Lett. 6, 024010. https://doi.org/10.1088/1748-9326/6/2/024010. Arima, E.Y., Barreto, P., Araújo, E., Soares-Filho, B., 2014. Public policies can reduce deforestation: lessons and challenges from Brazil. Land Use Policy 41, 465–473. https://doi.org/10.1016/j.landusepol.2014.06.026. Artaxo, P., 2019. Working together for amazonia. Science 363, 323. https://doi.org/10. 1126/science.aaw6986. Azevedo-Ramos, C., Moutinho, P., 2018. No man’s land in the Brazilian Amazon: Could undesignated public forests slow Amazon deforestation? Land Use Policy 73, 125–127. https://doi.org/10.1016/j.landusepol.2018.01.005. Barreto, P., Pinto, A., Brito, B., Hayashi, S., 2008. Quem é dono da Amazônia? Uma análise do recadastramento de imóveis rurais. Imazon, Belém. Begotti, R.A., Peres, C.A., 2019. Brazil’s indigenous lands under threat. Science 363, 6427. https://doi.org/10.1126/science.aaw3864. Borras Jr., S.M., Franco, J.C., 2010. Contemporary discourses and contestations around pro-poor land policies and land governance. J. Agrar. Change 10, 1–32. https://doi. org/10.1111/j.1471-0366.2009.00243.x. Campos-Silva, J.V., Peres, C.A., 2019. Brazil’s policies stuck in the mud. Science 363, 1046. https://doi.org/10.1126/science.aaw8293. Carvalho, W.D., Mustin, K., Hilário, R.R., Vasconcelos, I.M., Eilers, V., Fearnside, P.M., 2019. Deforestation control in the Brazilian Amazon: a conservation struggle being lost as agreements and regulations are subverted and bypassed. Perspect. Ecol. Conser. 17. https://doi.org/10.1016/j.pecon.2019.06.002. Deininger, K., 2011. Challenges posed by the new wave of farmland investment. J. Peasant Stud. 38, 217–247. https://doi.org/10.1080/03066150.2011.559007. Deininger, K., Feder, G., 2009. Land registration, governance, and development: evidence and implications for policy. World Bank Res. Obser. 24, 233–266. https://doi.org/10. 1093/wbro/lkp007. FAO – Food and Agriculture Organization of the United Nations, 2010. Global Forest Resources Assessment. Food and Agriculture Organization of the United Nations, Rome, LZ FAO Forestry Paper 163. Ferrante, L., Fearnside, P.M., 2019. Brazil’s new president and ‘ruralists’ threaten Amazonia’s environment, traditional peoples and the global climate. Environ. Conserv. 46. https://doi.org/10.1017/S0376892919000213. Garrett, R.D., Lambin, E.F., Naylor, R.L., 2012. Land institutions and supply chain configurations as determinants of soybean planted area and yields in Brazil. Land Use Policy 31, 358–396. https://doi.org/10.1016/j.landusepol.2012.08.002. Gibbs, H.K., Rausch, L., Munger, J., Schelly, I., Morton, D.C., Noojipady, P., Soares-Filho, B., Barreto, P., Micol, L., Walker, N.F., 2015. Brazil’s soy moratorium. Science 347, 377–378. https://doi.org/10.1126/science.aaa0181. Gibbs, H.K., Munger, J., L’Roe, J., Barreto, P., Pereira, R., Christie, M., Amaral, T., Walker, N.F., 2016. Did ranchers and slaughterhouses respond to zero-deforestation agreements in the Brazilian Amazon? Conserv. Lett. 9, 32–42. https://doi.org/10.1111/ conl.12175. Guedes, S.N.R., Reydon, B.P., 2012. Direitos de propriedade da terra rural no Brasil: uma proposta institucionalista para ampliar a governança fundiária. Rev. Econ. Sociol. Rural 50, 525–544. https://doi.org/10.1590/S0103-20032012000300008. Homma, A., 2010. Política agrícola ou política ambiental para resolver os problemas da Amazônia? Rev. Polít. Agríc. 19, 99–102. INCRA – Instituto Nacional de Colonização e Reforma Agrária, 1999. Livro branco da grilagem de terras no Brasil. Ministério da Política Fundiária e do Desenvolvimento Agrário, Brasília, DF. Lapola, D.M., Martinelli, L.A., Peres, C.A., Ometto, J.P.H.B., Ferreira, M.E., Nobre, C.A., Aguiar, A.P.D., Bustamante, M.M.C., Cardoso, M.F., Costa, M.H., Joly, C.A., Leite, C.C., Moutinho, P., Sampaio, G., Strassburg, B.B.N., Vieira, I.C.G., 2014. Pervasive transition of the Brazilian land-use system. Nat. Clim. Change 4, 27–35. https://doi. org/10.1038/nclimate2056. Laurance, W.F., Albernaz, A.K.M., Schroth, G., Fearnside, P.M., Bergen, S., Venticinque, E.M., Costa, C., 2002. Predictors of deforestation in the brazilian amazon. J. Biogeogr. 29, 737–748. https://doi.org/10.1046/j.1365-2699.2002.00721.x. Lima, M.C.M., 2002. Relatório das correições extraordinárias nos registros de terras rurais no Estado do Amazonas. Governo do Estado do Amazonas, Secretaria da Cultura do Estado do Amazonas, Manais, AM.

from the lack of adequate land administration and governance systems, and explored the ways in which insecure property rights, especially over public lands, result in land grabbing and deforestation. Section 7 drew upon the existing literature to demonstrate how effective land administration and governance is a prerequisite for reductions in deforestation. Section 8 showed the significant gains achieved by improvements in land administration and governance systems in Brazil, with a special focus on the newly created cadaster (CNI with SIGEF) and the initial successes of the Terra Legal public land regularization program. Lastly, the grim future scenario for environmental policy in Brazil was discussed, especially after the recent government measures against environmental protection and against indigenous people’s land rights. Bringing all of these pieces together, it is clear that, in order to combat deforestation in the Brazilian Amazon, it is necessary to move forward with improvements in the Land Administration/Governance System. The land cadaster must be completed and integrated with the CAR. The regularization of ownership, along the lines of the Terra Legal program, must be continued and expanded to include state-controlled public lands. Once a reliable land cadaster is in place, land taxes can be improved, which in turn diminish speculation and improve control over forested lands and associated environmental crimes. Finally, it is important to stress that, while better land governance is a necessary precondition for reducing deforestation, it is not in itself sufficient. Land governance and environmental protection must be structured under long term compromises and insulated against the politics of the day. Other policies, including Payment for the Environmental Services of the Forest (REDD+), development of markets for Non-Wood Forest Products, and more Command and Control policies undoubtedly have important roles to play in the struggle against deforestation of the Amazon rainforest. The close future scenario for deforestation in the Amazon is grim, especially given the newly elected government positions as anti-science and against the environmental conservation movements, which are already showing impacts as deforestation is increasing for the current year. 12

Land Use Policy xxx (xxxx) xxxx

B.P. Reydon, et al. Magalhães, A.L.B., 2017. Biodiversity at risk from austerity law. Nature 542, 295. https:// doi.org/10.1038/542295e. Margulis, S., 2001. Quem são os agentes do desmatamento na Amazônia e por que eles desmatam? Banco Mundial, Brasília, DF. Margulis, S., 2003. Causes of Deforestation of the Brazilian Amazon. The World Bank, Washington, DC World Bank working paper 22. Moutinho, P., Guerra, R., Azevedo-Ramos, C., 2016. Achieving zero deforestation in the Brazilian Amazon: what is missing? Elem Sci Anth 4, 000125. https://doi.org/10. 12952/journal.elementa.000125. Nepstad, D., McGrath, D., Stickler, C., Alencar, A., Azevedo, A., Swette, B., Bezerra, T., DiGiano, M., Shimada, J., Motta, R.S., Armijo, E., Castello, L., Brando, P., Matt, C.H., McGrath-Horn, M., Carvalho, O., Hess, L., 2014. Slowing Amazon deforestation through public policy and interventions in beef and soy supply chains. Science 344, 1118–1123. https://doi.org/10.1126/science.1248525. Oliveira, G.L.T., 2016. Regularização fundiária e a “corrida mundial por terras” no Brasil. Campo-Território: Rev. Geo. Agr. 11, 43–75. https://doi.org/10.14393/RCT112302. Ondetti, G., 2016. The social function of property, land rights and social welfare in Brazil. Land Use Policy 50, 29–37. https://doi.org/10.1016/j.landusepol.2015.08.028. Pacheco, P., 2009. Agrarian reform in the Brazilian Amazon: its implication for land distribution and deforestation. World Dev. 37, 1337–1347. https://doi.org/10.1016/ j.worlddev.2008.08.019. Paixao, S., Hespanha, P.J., Ghawana, T., Carneiro, A.F.T., Zevenbergen, J., Frederico, L.N., 2015. Modeling indigenous tribes’ land rights with ISO 19152 LADM: a case from Brazil. Land Use Policy 49, 587–597. https://doi.org/10.1016/j.landusepol. 2014.12.001. Palmer, D., Fricska, S., Wehrmann, B., 2009. Towards Improved Land Governance. FAO/ UN-Habitat. Land Tenure Working Paper 11. Pfaff, A., Robalino, J., Sandoval, C., Herrera, D., 2015. Protected area types, strategies and impacts in Brazil’s Amazon: public protected area strategies do not yield a consistent ranking of protected area types by impact. Philos. Trans. Biol. Sci. 370, 20140273. https://doi.org/10.1098/rstb.2014.0273. Petherick, A., 2017. Funding: austerity bites deeply. Nature 548, 249–251. https://doi. org/10.1038/nj7666-249a. Rausch, L.L., Gibbs, H.K., 2016. Property arrangements and soy governance in the Brazilian state of Mato Grosso: implications for deforestation-free production. Land 5, 7. https://doi.org/10.3390/land5020007. Reydon, B.P., 2019. Challenges of current land governance in Brazil: beyond the historical, political and social demands for land reform. In: Buainain, A.M., Lanna, R., Navarro, Z. (Eds.), Agricultural Development in Brazil: the Rise of a Global Agro-Food Power. Routledge, New York, NY, pp. 208–227 2019. Reydon, B.P., Fernandes, V.B., 2017. Financialization, land prices and Land Grab: study based on the Brazilian reality. Econ. Soc. 26, 1151–1181. https://doi.org/10.1590/ 1982-3533.2017v26n4art12. Reydon, B.P., Telles, T.S., Fernandes, V.B., 2015. Land tenure in Brazil: the question of regulation and governance. Land Use Policy 42, 509–516. https://doi.org/10.1016/j. landusepol.2014.09.007.

Reyes-Galindo, L., Monteiro, M., Macnaghten, P., 2019. “Opening up” science policy: engaging with RRI in Brazil. J. Responsible Innov. 6. https://doi.org/10.1080/ 23299460.2019.1603568. Richards, P.D., Walker, R.T., Arima, E.Y., 2014. Spatially complex land change: the Indirect effect of Brazil’s agricultural sector on land use in Amazonia. Glob. Environ. Change 29, 1–9. https://doi.org/10.1016/j.gloenvcha.2014.06.011. Robinson, B.E., Holland, M.B., Naughton-Treves, L., 2014. Does secure land tenure save forests? A meta-analysis of the relationship between land tenure and tropical deforestation. Glob. Environ.l Change 29, 281–293. https://doi.org/10.1016/j.gloenvcha. 2013.05.012. Shiki, S., Shiki, S.F.N., 2011. Os desafios de uma política nacional de pagamentos por serviços ambientais: lições a partir do caso do Proambiente. Sustentabilidade em Debate 2, 99–118. https://doi.org/10.18472/SustDeb.v2n1.2011.3909. Silva, L.O., 1996. Terras devolutas e latifúndio: efeitos da lei de 1850. Unicamp, Campinas, SP. Soares-Filho, B., Rajão, R., Macedo, M., Carneiro, A., Costa, W., Coe, M., Rodrigues, H., Alencar, A., 2014. Cracking Brazil’s forest code. Science 344, 363–364. https://doi. org/10.1126/science.1246663. Soares-Filho, B.S., Nepstad, D.C., Curran, L., Cerqueira, G.C., Garcia, R.A., Ramos, C.A., Voll, E., McDonald, A., Lefebvre, P., Schlesinger, P., McGrath, D., 2005. Cenários de desmatamento para a Amazônia. Estud. Av. 19, 137–152. https://doi.org/10.1590/ S0103-40142005000200008. Sonter, L.J., Herrera, D., Barrett, D.J., Galford, G.L., Moran, C.J., Soares-Filho, B.S., 2017. Mining drives extensive deforestation in the Brazilian Amazon. Nat. Commun. 8, 1013. https://doi.org/10.1038/s41467-017-00557-w. Sparovek, G., Berndes, G., Barretto, A.G.O.P., Klug, I.L.F., 2012. The revision of the Brazilian Forest Act: increased deforestation or a historic step towards balancing agricultural development and nature conservation? Environ. Sci. Policy 16, 65–72. https://doi.org/10.1016/j.envsci.2011.10.008. Sparovek, G., Reydon, B.P., Pinto, L.F.G., Faria, V., Freitas, F.L.M., Azevedo-Ramos, C., Gardner, T., Hamamura, C., Rajão, R., Cerignoni, F., Siqueira, G.P., Carvalho, T., Alencar, A., Ribeiro, V., 2019. Who owns Brazilian lands? Land Use Policy 87, 104062. https://doi.org/10.1016/j.landusepol.2019.104062. Stickler, C.M., Nepstad, D.C., Azevedo, A.A., McGrath, D.G., 2013. Defending public interests in private lands: compliance, costs and potential environmental consequences of the Brazilian Forest Code in Mato Grosso. Philos. Trans. Biol. Sci. 368, 20120160. https://doi.org/10.1098/rstb.2012.0160. Telles, T.S., Palludeto, A.W.A., Reydon, B.P., 2016. Price movement in the Brazilian land market (1994-2010): an analysis in the light of post-Keynesian theory. Rev. Econ. Polit. 36, 109–129. https://doi.org/10.1590/0101-31572016v36n01a07. Telles, T.S., Reydon, B.P., Fernandes, V.B., 2018. Os determinantes do preço das terras agrícolas na história do pensamento econômico. Econ. Soc. 27, 525–545. https://doi. org/10.1590/1982-3533.2017v27n2art6. Wilkinson, J., Reydon, B., Sabbato, A., 2012. Concentration and foreign ownership of land in Brazil in the context of global land grabbing. Can. J. Dev. Stud. 33, 417–438. https://doi.org/10.1080/02255189.2012.746651.

13