The Representational Effects of Communal Property: Evidence from Peru’s Indigenous Groups

Abstract

Why do some indigenous groups achieve coethnic political representation while others do not? In this paper, I highlight the primary role of communal property in shaping indigenous representation. While scholars often laud the developmental benefits of communal land titling, I argue that formalizing collectively held land can inhibit indigenous coordination to achieve political representation. Where communal land is informally held, indigenous groups are more likely to invest in traditional institutions that facilitate collective action to elect coethnic candidates to political office. Conversely, titling communal property secures indigenous land access but in the process erodes traditional institutions that would otherwise promote collective action during elections. I test my argument using a multi-method approach that includes interviews and experiments with three-hundred Peruvian indigenous leaders, historical land-title data, and information scraped from mayoral candidate CVs. The findings suggest that the oft-cited economic benefits of collective property may generate negative political effects.

Keywords

indigenous politics representation and electoral systems race ethnicity and politics Latin American politics

Indigenous groups throughout the Americas have historically experienced political exclusion and underrepresentation. Yet, in recent decades, decentralization reforms, electoral quotas, and ethnically based political parties have created institutional spaces through which indigenous groups can achieve coethnic political representation at various levels of government (Hoffay & Rivas, 2016; Htun, 2016; Madrid, 2012; Rice & Van Cott, 2016; Van Cott, 2009). While some indigenous communities have effectively coordinated to take advantage of these opportunities for greater representation, others have been less successful. In this paper, I analyze the sources of this variation. Specifically, why have some indigenous groups successfully mobilized to achieve coethnic political representation while others have failed to do so?

To explain variation in indigenous groups’ ability to achieve coethnic political representation, I focus on a key institutional reform that has often defined indigenous-state relations over the past century: the titling of native groups’ communal land.¹ Indigenous groups possess the majority of the world’s communal land, and for these groups, the titling of communal land constitutes an essential step toward achieving their “central demand” of greater autonomy (Díaz-Polanco, 1998; Stavenhagen, 2000; Van Cott, 2001; Yashar, 2005). State recognition and formalization of communal property also provides indigenous groups the promise of economic benefits, which are generally thought to equal or even exceed those of private property (Cramb & Wills, 1990; Sjaastad & Bromley, 1997; cf. Place & Hazell, 1993). Unlike private property, a communal land title belongs to no single individual but rather to the “community” that occupies the land, and this collectively held property—by law—is generally indivisible and inalienable.

I argue that the purported economic benefits of communal land titles can be accompanied by negative political effects, particularly with respect to indigenous representation. Communal land titles allow indigenous groups to shift their focus from defending their land to improving agricultural productivity and integrating into markets. While this transition may provide certain private economic benefits, it may also erode traditional institutions of reciprocity (Polanyi, 1944). The loss of these institutions complicates intra-group electoral coordination to achieve coethnic political representation.

Informally held communal land, on the other hand, is more likely to encourage the persistence of reciprocity institutions, which promote indigenous groups’ electoral coordination in two ways. First, indigenous candidates who subscribe to these institutions can more credibly commit to reciprocating electoral support with post-election benefits if they win. Therefore, these candidates will be generally more appealing to indigenous voters. Second, reciprocity institutions facilitate turn-taking among communities around who nominates candidates for local office. Voters from a given indigenous community will support a candidate from another community, confident that a member of their community will have an opportunity to run for office and will receive similar support in a future election.

I examine this argument drawing on evidence from Peru, where ethnicity is a fairly weak political identity (Albó, 2008; 1991; Degregori, 1998; Yashar, 2005).² As such, political coordination across indigenous communities cannot be easily explained by ethnic solidarity or primordial ties, thus facilitating an analysis of other factors that may affect inter-community electoral mobilization. Crucially, the Peruvian case also exhibits variation in the key independent variable of interest to this study. Peru features both formally titled and informally possessed communal landholding institutions, or “communities.” This feature of the Peruvian case allows me to trace how different forms of communal landholding—titled or informally held—have shaped the persistence of traditional institutions and ultimately, political representation.

I use a multi-method approach that draws on historical and contemporary evidence. I first show that the 20th-century extension of communal land titles was associated with an increase in market integration and a decline in reciprocity institutions, or ayni. As further evidence that reciprocity institutions were less likely to persist in title-possessing communities, I draw on a lab-in-the-field experiment with over three-hundred indigenous community presidents. I show that reciprocal behavior is more likely to be observed in communities that historically lacked a formal title to their communal land.

A second empirical section demonstrates the important coordinating role of traditional reciprocity institutions in achieving coethnic representation. First, I use data scraped from the CVs of candidates for municipal office in 2014 to show that reciprocity institutions reduce the number of community-member candidates who run in a given municipality. Then, using a conjoint experiment, I provide evidence that cross-community coordination to elect indigenous candidates to subnational office is more likely to occur in communities that preserve reciprocity institutions. Finally, I draw on interview and case study evidence from two Peruvian provinces to illustrate how indigenous communities coordinate to achieve coethnic political representation.

The findings offer insights into not only indigenous groups’ access to representation, but also their access to state-provided distributive benefits. Scholars have often observed that the scarcity of government investment in indigenous communities is driven by low levels of indigenous representation, especially in local governments (Freire et al., 2015; Hoffay & Rivas, 2016; Htun, 2016; Van Cott, 2005). In other work, I use a regression-discontinuity design to show that indigenous groups receive more public goods when municipal leaders are themselves indigenous (Carter, 2021a). Therefore, an increase in descriptive representation also corresponds to an improvement in indigenous groups’ substantive representation.

Theoretical Framework

Communal land titles have been one of the most frequent policy concessions governments have offered to indigenous groups. As Figure 1 illustrates, formally titled indigenous communal lands cover a large share of contemporary Andean South America and Central America. These include comunidades indígenas (indigenous communities),³ resguardos (reserves), and comarcas (regions).⁴ However, also notable from Figure 1 are the many areas of relatively high indigenous population where communal land has not been titled, like southern Mexico, Guatemala, Honduras, highland Ecuador, northern Chile, and central Bolivia. In these areas, indigenous groups often occupy communal land informally and must engage in constant mobilization to protect and preserve their territorial integrity.

Figure 1.

Indigenous communal land (titled, 2015).

In this paper, I argue that the titling of communal land reduces indigenous groups’ access to political representation. It does so by encouraging market integration, a process that erodes traditional institutions of reciprocity that would otherwise facilitate coethnic coordination during elections.

In the absence of a formal title, indigenous groups must remain vigilant in guarding their land against external threats. As individuals or a group, community members may encounter non-indigenous farmers, miners, or settlers who desire their land. Legally enforced communal titles remove the need for indigenous peasants to defend their land from external encroachment. Assuming a fixed level of human resources, land tenure security allows indigenous households to devote more time and resources to commercial agriculture.⁵

In turn, this greater market participation reduces the demand for traditional reciprocity institutions, which have long structured the social, political, and economic life of native tribes, kinship groups, and communities in the Americas.⁶ The most prevalent of these institutions involves an individual-to-individual transfer in which one community member performs a service for another—such as harvesting crops or defending another member’s land against invasion from outsiders—with the agreement that the recipient will offer the provider a similarly useful form of future assistance. Examples include ayni in Peru, Ecuador, and Bolivia; ké’í among the Navajo in the United States (Austin, 2009, pp. 145–146); and mano vuelta in Mexico and Central America.

Market integration generally undermines these reciprocity institutions.⁷ First, participation in commercial agriculture makes peasants more responsive to cash payments and a profit motive than to reciprocal exchanges of unpaid labor (Migdal, 1974; Paige, 1978; Scott, 1977; Thompson, 1971).⁸ Guillet (1980), for example, observes that reciprocity institutions will survive only where “market forces have not yet penetrated sufficiently to make wage labor an efficient and available alternative” (p. 158). Second, market integration promotes new forms of socioeconomic differentiation.⁹ Subsequently, economically powerful leaders in legally recognized communities may exploit poorer community members, including through reciprocity institutions (Winder, 2014, p. 232). These dynamics have increased the reluctance of many community members to participate in reciprocity institutions, thereby leading to their erosion. Importantly, this decline of reciprocity institutions—and the market integration that proceeds it—does not occur immediately but rather gradually, often over the course of decades.¹⁰

The erosion of traditional reciprocity institutions has important implications for indigenous political representation. While these institutions are often preserved within individual communities, they also serve a critical function in facilitating coethnic electoral coordination across communities (Habyarimana et al., 2009). First, they may limit the number of indigenous candidates who run for office in a given jurisdiction (e.g., municipality). Absent reciprocity institutions, candidates from multiple communities in a given municipality may run for office. The existence of reciprocity institutions facilitates turn-taking among these communities. Indigenous communities often decide among themselves that a candidate from a given community will run for elective office at the local level. Voters from other communities within that electoral circumscription support the chosen candidate, confident that a member of their own community will be selected to run in the future. When that person runs, he or she will likewise have the support of other indigenous communities. Without reciprocity institutions, communities could not credibly commit to such turn-taking. In this way, these longstanding institutions—often translated from indigenous languages as “today for you, tomorrow for me”—play a key role in limiting the supply of indigenous candidates who may run and thereby prevent a division of the indigenous vote among many coethnic candidates for elective office.

Reciprocity institutions also facilitate electoral coordination across indigenous communities in a second way. A key challenge for political candidates in electoral democracies involves their inability to credibly commit to fulfilling their campaign promises (Keefer & Vlaicu, 2008). Voters often do not know how much—if at all—they can trust individual candidates to comply with their stated platforms. While voters may be able to trust candidates from their own community, they generally trust outsiders less. The preservation of reciprocity institutions, however, provides a commitment device. Specifically, candidates from indigenous communities where reciprocity institutions are preserved can credibly commit to reciprocating votes with post-election distributive benefits. Importantly, this commitment device operates even among candidates and voters who are not from the same community.

The existence of reciprocity institutions thus generates preferences among indigenous voters for candidates who adhere to these institutions. In addition to reducing the supply of indigenous candidates, reciprocity institutions also coordinate indigenous voters’ support around common coethnic candidates for elective office.

My theory therefore predicts that informally held communal land will increase indigenous groups’ access to political representation. These landholding arrangements thwart market integration and thereby preserve traditional institutions of reciprocity, which help native groups overcome coordination problems to mobilize around common coethnic candidates for local office. Communal land titles, on the other hand, encourage capitalist production in communities and erode traditional institutions of reciprocity by removing a shared interest in survival that otherwise binds groups together. This loss of reciprocity institutions complicates indigenous groups’ ability to achieve political representation. I outline the argument and evidence used to evaluate it in Figure 2. The rest of the article tests this theory using the case of Peru’s indigenous comunidades, or “communities.” While certain pieces of evidence allow for more robust causal inference than others, the cumulation of multi-method data lends support to the theory presented above.

Figure 2.

Communal land titles and political representation: theory and evidence.

Indigenous Communal Landholding in Peru

In Peru, longstanding institutions of indigenous authority play an important—but largely understudied—role in shaping political, economic, and social life in rural areas.¹¹ Chief among these institutions are indigenous communities, which are parcels of communally held land that are administered—to varying degrees—by longstanding authorities. Communities are the “oldest institution of [Peruvian] society” (Mendoza, 2002, p. 8), and, while they have been formally called “peasant communities” since the 1960s, they were initially titled as “indigenous communities” and today are almost exclusively populated by indigenous citizens. By “indigenous,” I mean individuals who maintain longstanding customs and who speak an indigenous language, such as Quechua or Aymara.

Indigenous communities remain prevalent throughout Peru, controlling around 30% of national territory and accounting for roughly 20% of the country’s population.¹² In total, there are over 7,000 indigenous communities, and all but around 15% have now been formally recognized.

These communities are generally characterized by low-quality soil, unreliable irrigation, limited access to markets, and widespread poverty (Webb, 2013).¹³ Many are located in Peru’s mountainous Andean sierra, particularly the southern departments of Puno, Cusco, Arequipa, and Ayacucho. Communities are nested within municipalities—Peru’s lowest administrative tier of government—and often there exist several communities within a given municipality, particularly in the sierra.

Many of Peru’s indigenous communities continue to maintain reciprocity institutions, particularly through ayni (Mannheim, 2013, pp. 89–90). Ayni, which means reciprocity in the indigenous languages of Quechua and Aymara (Stern, 1993, p. 8), obligates community members to make a contribution of labor, or minka/mita, to their fellow community members with the expectation that that favor, in some way, will be returned (Fonseca Martel, 1974a, 1974b; Mayer, 2018; Wutich et al., 2017). Unlike communal work parties to produce public works for the community, or faenas, ayni involves a person-to-person or household-to-household exchange. For example, a male member of a community may ask others to step in to help him harvest his crops, make needed repairs to his home, or defend his land against invasion by outsiders. Later, others may—and likely will—call on this community member to assist with their labor-intensive tasks.

These institutions are maintained through a complex enforcement system that includes both social sanctioning and explicit punishments (Estermann, 1998, p. 250). With respect to the latter, non-participation in an ayni task may be punished through the levying of a fine, imprisonment in a community jail, public flogging, or—for serial offenders—banishment from the community (Mannheim, 2013, p. 90). Often, the community president or a representative explicitly in charge of communal justice will supervise the punishment.¹⁴

While reciprocity institutions persist in most indigenous communities, there has been a notable decline in their mobilization, particularly in those communities most affected by market integration. In my interviews with indigenous community presidents, I was frequently told that the greatest barrier to mobilizing reciprocity institutions in the modern period is the cash economy; if a member has an obligation to provide unpaid labor to the community or one of its members, she prefers to pay her way out of that commitment rather than working for free.¹⁵ Those who request ayni for tasks like harvesting crops or building a house also increasingly prefer wage labor to unpaid labor; payment of a wage offers the contracting community member a way of both holding workers accountable and ensuring a more timely completion of the desired task (Seligmann, 1995, pp. 141–142).

Some indigenous communities have experienced a greater reduction in their mobilization of reciprocity institutions than others. What accounts for this decline? As I show in the next section, variation in possession of a communal land title has played a key role in determining the fate of these longstanding institutions.

Communal Landholding and Reciprocity Institutions

Reciprocity institutions operate most effectively when community members prioritize group survival over individual profit. I argue above that informal possession of communal land requires community members to cooperate to ensure the joint survival of their land and their institutions. Communal titles, on the other hand, provide greater security for indigenous communal land, which erodes the joint interest in survival that community members otherwise possess and thereby, displaces traditional reciprocity institutions. This section explores the effects of these different communal landholding arrangements on longstanding institutions of reciprocity.

Observational Analysis on Historical Data

In 1925, the Peruvian government issued indigenous communal land titles for the first time in the post-independence period. Over the next 80 years, communities steadily obtained titles for their communal land, and by 2007, nearly 80% of indigenous communities had obtained a formal title. The theory predicts that these titles should have resulted in greater market integration and in turn eroded reciprocity institutions. Therefore, the longer a community has a title the more likely it should be to experience market integration and the less likely it should be to preserve traditional institutions of reciprocity. To test this, I gather data on the year that communities received a formal title—from a 2009 community directory (Ministerio de Vivienda, Construccion, y Saneamiento, 2009)—and measures of market integration and reciprocity institutions, gathered from a 2012 census of indigenous peasant communities (Instituto Nacional de Estadística e Informática, 2014).

Figure 3 provides support for the first steps in the theory.¹⁶ The first panel plots the bivariate relationship between the years since a community obtained its title (from 2009) and a community’s likelihood of engaging in any number of productive and business activities, including agriculture, livestock, mining, forestry, and tourism. There is a relatively strong positive correlation, suggesting that communities that received their titles earlier are more likely to engage in economic activities oriented toward external markets today. The second panel investigates whether communal land titles also erode traditional institutions of reciprocity. The bivariate analysis suggests a negative association between the length of time a community has possessed a title and its likelihood of preserving reciprocity institutions (i.e., ayni). Finally, there appears to be a negative relationship between market integration and the maintenance of reciprocity institutions. A regression of the aforementioned dummy reciprocity measure on the dichotomous measure of market integration yields a strong, negative association between the two variables ( $F = 21.74$ ).¹⁷

Figure 3.

Over-time correlation between communal land title, market integration, and reciprocity institutions: (a) likelihood of market integration and (b) preservation of reciprocity institutions (ayni).

Figure 4 plots the relationship between length of time since obtaining a title and a community-level additive index that sums the aforementioned reciprocity and market-integration variables, where the latter is recoded as “1” if there is no market integration and “0” if there is market integration.¹⁸ The fit remains strong and in the expected negative direction ( $r = - 0.67$ ).

Figure 4.

Over-time correlation between communal land title and combined measure of market integration and reciprocity institutions.

One potential concern arising from this analysis involves the potential endogeneity of market integration and communal land titles. For example, communities that were already integrated into markets may have also been more likely to seek a communal title because, for instance, they had the financial resources to pay fees to land surveyors. Yet, this seems unlikely because the requirements and thus the cost of receiving a title have changed substantially over time in ways that do not appear to correspond to the evidence presented above. Throughout the 1920s, surveyors charged exorbitant rates to communities to produce cadastral maps. In 1931, the government fixed the rate surveyors could charge at two centavos, far less than the average rate communities were paying previously (Davies, 1974). Then, in 1961, the Peruvian government increased the number of documents communities had to produce, raising anew the cost of obtaining a communal title. If market integration determined the date that communities received a title, a more volatile relationship should emerge in Figure 3a with discontinuous jumps in the years where policies reduced or increased the cost of a title. Yet, Figure 3a shows a relatively steady upward trend.

A second potential concern arising from the above analysis involves alternative explanations that are unaccounted for by the bivariate correlation. For example, Quechua-speaking communities, which traditionally had somewhat greater access to the central state, may have been better positioned to receive an early title than Aymara-speaking communities; different ethno-linguistic communities may also have different experiences with markets and reciprocity institutions. Geographic location presents another factor that may have influenced both the date of title as well as the two dependent variables. Communities near the capital of Lima, for instance, may have obtained titles earlier and may have also been more integrated into markets and less likely to preserve reciprocity institutions.

In Table 1, I present an OLS regression of the market-ayni index from Figure 4 on the covariates for which data are available at the community-level and that pre-date the issuing of the first community title in 1925. In addition to the main independent variable—years since title—I include dummy indicators for Quechua-speaking communities, Aymara-speaking communities, and the existence of a colonial-era communal title. I also include municipal fixed effects to account for other time-invariant factors operating at the municipal level. While the estimates presented in the table do not carry a causal interpretation, they lend further support to the existence of a strong negative association between the length of communal title and the market-ayni index even when accounting for other variables.¹⁹

Table 1.

OLS Regression of Market-Ayni Index on Years Since Title and Covariates.

	Dependent variable
	Market-ayni index
Years since title	−0.003*** (0.001)
Quechua-speaking	−0.238 (0.519)
Aymara-speaking	−0.033 (0.521)
Colonial title	0.034 (0.070)
Observations	2,651
Municipal FEs	Yes
R ²	0.041
Adjusted R²	0.029
Residual std. error	0.661 (df = 2619)

Market-ayni index is a community-level variable that sums dummy variables for ayni and market integration and takes on values: 0, 1, or 2. 0 indicates market integration and no ayni, while 2 indicates ayni and no market integration. See Figure 3 for sources. Colonial title is a dummy variable taken from 2012 Cenagro Census. Quechua-speaking and Aymara-speaking are also dummy variables indicating whether a majority of community members speak Quechua or Aymara, respectively, as opposed to another indigenous language.

Robust SEs in parentheses *p< .1. **p < .05. ***p < .01.

Communal Titles and Reciprocity Institutions: Lab-in-the-Field Experiment

As a further test of the relationship between communal titles and reciprocity institutions, I conduct an experiment with over 300 indigenous community presidents in Cusco. A description of this sample is provided in the Supplemental Appendix.

Community presidents are officially recognized by the Peruvian state, but they are not formal agents of the state, as mayors and governors are. Instead, community presidents are popularly elected by their members in a manner that is codified in each community’s constitution. Election may occur by secret ballot, voice vote, or debate in a public assembly. Once selected, presidents serve 1 to 3 year terms. During that time, they are responsible for maintaining and administering traditional practices, including reciprocity institutions. Thus, measuring whether presidents themselves subscribe to reciprocity institutions provides valuable insight into whether their communities preserve these institutions.

To measure reciprocity, I employ a trust game first proposed by Berg et al. (1995). As scholars have noted, the first-mover’s behavior in a trust game measures two key dimensions of reciprocity: “intrinsic reciprocity”—eliciting kindness through kindness regardless of the monetary outcome—and expected reciprocity, which includes a more “selfish” calculation based only on monetary payouts (Brülhart & Usunier, 2012; Sobel, 2005). Thus, the experimental game I use provides a behavioral measure of reciprocity institutions to accompany the attitudinal outcome measure employed in the previous section.

In the trust game, presidents were, at the outset, given three Peruvian soles (approximately US$1), slightly more than an average hourly wage in rural Peru.²⁰ This amount could be either kept or shared with a partner. Presidents were informed that any money given to their partner would be tripled. Presidents were further informed that their partner would then—as in a traditional dictator game—decide how much money they would return to the president. A measure of trust, or in this case reciprocity, is thus how much of the three soles each president allocated to a partner.²¹

Community presidents were randomly assigned to receive certain information about their partner. Approximately half of the respondents were told that their partner was from an indigenous community in Cusco while the other half were told simply that their partner was from Cusco.²² Reciprocity institutions are generally maintained only among members of indigenous communities. Therefore, respondents should be more likely to expect reciprocal behavior when informed that their partner is from an indigenous community. Larger gifts to partners from indigenous communities thus indicate stronger reciprocity institutions.

A potential concern is that telling a respondent that their partner is from an indigenous community may signal that the partner is poor, and therefore, the game is measuring the altruism of the respondent as opposed to reciprocity. However, existing scholarship shows that altruism does not explain “trust-like decisions” in trust games (Brülhart & Usunier, 2012).

An examination of key pre-treatment covariates for the treatment and control groups shows balance, as expected due to the randomization of the treatment.²³ For only one of the baseline characteristics is there a significant difference between treated and control groups: whether the respondent is a current or former community president. While there is a higher proportion of current presidents in the treatment group than in the control group, this variable is correlated only weakly with the outcome ( $r = 0.02$ ), and its inclusion as a moderator neither substantively nor significantly changes the results.²⁴

Using the results of this game, difference in means were estimated on two different experimental outcomes. The first outcome is the amount given by respondents to their partner, which can take on integer values, 0 to 3. The second outcome is the probability that the president gave something, as opposed to nothing, coded dichotomously as 0 or 1 for each respondent.²⁵

Table 2 provides the main results from the experimental game using a regression of the outcome on a dichotomous variable indicating treatment assignment. On average, respondents gave just under half of their allotment (≈1.5 soles) to their partners. When informed that their partner was from a community, respondents gave around 0.25 soles more than when they were told only that their partner was from Cusco.²⁶ Using a dummy measure of whether the respondent gave either something or nothing to their partner, I also find that respondents were more likely to give when their partner was identified as a community member. Being assigned to play with someone identified as a community member increased the likelihood of giving something by nearly 12 percentage points. These results not only hold but become stronger when analyzing the results as a complier average treatment effect, where compliance is coded using a manipulation check question that measures whether respondents remembered that their partner was a community member.²⁷ Ultimately, these results suggest that there generally exist shared norms of reciprocity among presidents of indigenous communities.

Table 2.

Effect of Community Membership on Giving Behavior.

	Dependent variable
	Absolute amount of gift		Binary measure of giving
	ITT	CACE	ITT	CACE
Partner is community member	0.246** (0.115)	0.577*** (0.171)	0.117** (0.047)	0.238*** (0.070)
Constant	1.302*** (0.084)	1.130*** (0.108)	0.705*** (0.034)	0.643*** (0.044)
Observations	317	316	317	316
Residual std. error	1.022	1.035	0.420	0.424

p < .1. **p < .05. ***p < .01.

Do the behaviors in the game, in fact, correspond to traditional norms of reciprocity? The game is designed to detect whether players expect their partner to reciprocate an act of generosity with a similar act—or at least an even division of resources. To test whether this behavior maps onto traditional reciprocity institutions, I perform a heterogeneous treatment effects analysis to see if reciprocal behavior is more likely in communities that report preserving ayni in the 2012 census of indigenous communities (Instituto Nacional de Estadística e Informática, 2014). Figure 5 measures the effect of a community-member partner on the amount given by community presidents, disaggregated by self-reported measures of reciprocity institutions. For presidents of communities that preserve ayni, the treatment effect is in the expected positive direction and significant ( $p < 0.05)$ . For presidents of communities that do not report preserving ayni, the result is negative and not statistically significant. These results thus suggest that the behavioral evidence around reciprocity institutions obtained from the lab-in-the-field is consistent with attitudinal measures obtained from earlier surveys of indigenous communities.

Figure 5.

Correspondence between behavioral and attitudinal measures of reciprocity.

The theory predicts that traditional reciprocity institutions should be less likely to persist under communal land titles—as shown in the preceding section—and more likely to persist where land is held informally. Figure 6 provides a test of this claim by analyzing heterogeneity in the lab-in-the-field experiment by communities’ experience with communal land titles. I compare early-adopters (pre-1960) with later adopters (post-1960) and those that have not received a formal title to their communal land.²⁸

Figure 6.

Heterogeneity in treatment effects by communal title adoption: (a) amount of gift and (b) probability of giving.

The results are generally consistent with my theory. For early-adopters, the treatment effect is not significant, suggesting no difference in levels of reciprocity between members of indigenous communities. Presidents of these communities gave, on average, half a sol less to partners who were identified as belonging to indigenous communities. Evidence of reciprocity is more pronounced among later adopters, although the difference between early and late adopters is not significant. For the few communities in the sample that continue to hold their land informally, there is much stronger evidence of the presence of reciprocity institutions, using both continuous and dichotomous measures of giving.²⁹ These presidents gave about 1.2 soles more to partners identified as belonging to an indigenous community and were about 50% more likely to give something—as opposed to nothing.

Thus, the results of the correlational analysis and the lab-in-the-field experiment suggest that traditional institutions of reciprocity are most likely to persist where communal land is held informally. In contrast, communal titles appear to undermine these longstanding reciprocity institutions. Importantly, these associations do not demonstrate causality, as observed and unobserved factors that determine date of title may also affect both market integration and the persistence of reciprocity institutions. Yet, through process tracing within individual communities, historians have observed similar relationships to those that emerge in the above data.

In a study of indigenous communities in the central highlands of Peru, Mallon argues that “the legal recognition of Indian communities would in the long run accomplish more than any other law to integrate the peasantry into the developing capitalist economy” (Mallon, 2014, p. 232). Consistent with the theory I propose, Mallon attributes this relationship primarily to changing patterns in how communities defended their land, as they shifted from self-defense to greater dependence on the enforcement of communal titles by the state (Mallon, 2014, p. 232).

In a separate study of two peasant communities in the same central highland region of Peru, Winder (2014) also shows how communal titling has increased market integration. He argues,

Since the comunidades of Matahausi and Sicaya were legally recognized. . . there has been little effort to establish joint production activities. . . parallel and often competing “cooperative” organizations have been established, and through them the more commercially oriented farmers have been encouraged to. . . cater to expanding markets (Winder, 2014, pp. 232–233).

Winder proceeds to show that titled communities have also witnessed a decline in traditional institutions of reciprocity, observing that “institutional systems for organizing labor exchanges and work parties, such as. . . ayni, are now of relatively minor importance” (Winder, 2014, p. 232). Nuñez Palomino (1996) more directly notes the connection between market integration and reciprocity institutions in newly recognized Peruvian indigenous communities, arguing that “the introduction of market values in the communal space. . . tends to minimize the role of [reciprocity] principles” (31).

Thus, communal titles appear to facilitate market integration and thereby reduce the persistence of reciprocity institutions. In the next section, I use quantitative and qualitative data to examine how these reciprocity institutions, when they persist, facilitate coordination across indigenous communities to elect coethnic candidates to subnational office.

Reciprocity Institutions and Descriptive Representation

I argue above that the preservation or erosion of reciprocity institutions has important representational effects. Specifically, reciprocity institutions generally facilitate coordination across communities to elect indigenous candidates to subnational office. These institutions strengthen preferences for candidates who are from any indigenous community—regardless of whether they are from the same community as a given indigenous voter. In this section, I evaluate these claims, using observational data from municipal candidate CVs, a conjoint experiment conducted with community presidents, and two case studies from fieldwork in Cusco, Peru.

I first use electoral and community census data to test whether reciprocity institutions facilitate inter-community electoral coordination within municipalities. I scraped the CVs of all candidates who ran for municipal president (i.e., mayor) in 2014, which yielded a dataset of over 11,000 candidates. I recorded each candidate’s place of residence and used this information to code whether each candidate was from an indigenous peasant community. I then aggregated this data up to the municipal level to construct a measure of how many community-member candidates ran in a given municipality. I include only municipalities where there exists more than one indigenous community and where at least one community member ran for municipal office in 2014.

Figure 7 provides evidence of a negative association between the preservation of ayni and the number of community-member candidates running for office. Where all communities in a municipality preserve ayni, there are fewer community-member candidates who run for local office. There is also a higher likelihood that only one such candidate will run. While the analysis does not carry a causal interpretation, it does lend support to a basic bivariate relationship of interest: greater electoral coordination appears to emerge in municipalities where reciprocity institutions are universally preserved.

Figure 7.

Association between municipal-level preservation of reciprocity institutions and electoral coordination (Peru, 2014): (a) likelihood of multiple candidates in municipality and (b) number of candidates in municipality.

I now turn to evidence from a conjoint analysis in which I use the same sample of indigenous community presidents from the lab-in-the-field experiment. In subnational elections, individual indigenous communities often decide internally whom they will support (Flórez, 2012). Community elites, including current and former community presidents, play an important role in this decision-making process, both by disseminating information about candidates and shaping opinions of community members. Further, community presidents are also community members. Thus, studying the electoral choices of presidents may provide insight into the broader political behavior of the community.

When administering the conjoint, enumerators first reminded presidents of the upcoming 2016 municipal elections. Presidents were then provided with five pairs of hypothetical candidate profiles and asked to both (1) pick which of the two candidates for whom they would vote and (2) rate their likelihood of voting for each of the candidates. The latter comprised a five-point scale, where “1” indicated very unlikely and “5” indicated very likely to support the candidate.

Only four attributes were provided for each candidate to reduce cognitive load for respondents. Attributes included the candidates’ gender, policy platform, party affiliation, and community membership. Through earlier interviews with community presidents, these attributes were determined to be the most important predictors of vote choice.

The main treatment of interest is whether a given mayoral candidate is a member of a community. Qualitative fieldwork and interviews with community presidents conducted prior to the administration of this conjoint demonstrated a shared understanding of the word “community” to specifically connote an indigenous community. Secondary scholarship also describes the linkage between the word “community” and notions of indigenous territory in Peru (Hurtado, 2012; Remy, 2013).

Respondents were informed that candidates were from their community, another community, or the district capital. To determine the relative effect of candidates’ community membership on presidents’ vote choice, I follow Hainmueller et al. (2014) and estimate an average marginal component effect (AMCE). I use a non-parametric estimation strategy for both the discrete choice- and rating-based outcomes. I estimate the following two models.

{rating}_{i j k} = θ_{0} + θ_{1} [{community}_{i j k} = 1] + θ_{2} [{community}_{i j k} = 2] + ϵ_{i j k}

(1)

{choice}_{i j k} = θ_{0} + θ_{1} [{community}_{i j k} = 1] + θ_{2} [{community}_{i j k} = 2] + ϵ_{i j k}

(2)

where ${rating}_{i j k}$ is the rating assigned by respondent i to profile j on task k. The reference category—a candidate from the district capital—provides a baseline and will be captured by $θ_{0}$ . θ₁ and θ₂ represent the coefficients for whether a candidate is from the respondent’s community or another community, respectively. Standard errors are clustered at the individual level (Hainmueller et al., 2014, p. 17). I also estimate models that include all attributes as a way of assessing the relative weight of each of the attributes.³⁰

Figures 8 and 9 offer support for a central hypothesis of this paper: that a member of a community—whether it is the president’s community or another one—is a strong and significant positive predictor of vote choice.³¹ Receiving information that a candidate was from the same community increased the likelihood that the president supported that candidate by 20 percentage points over a candidate identified as being from the district capital. This information also led to a nearly half-point increase in support for the candidate on the five-point scale.

Figure 8.

Conjoint analysis: predictors of candidate rating.

Figure 9.

Conjoint analysis: predictors of candidate choice.

Additionally, the results suggest a more general positive effect of community membership. Receiving information that the candidate was from a different community had a strong, positive effect on community presidents’ vote choice. Compared to candidates from the district capital, candidates from different communities than the respondent were 10 percentage points more likely to receive support on the discrete outcome measure and received an extra third of a point on the five-point rating outcome.³² As expected, due to the randomization of attribute levels, these results remain unchanged when covariates for attribute levels are included in the regression of the outcome on candidates’ community membership.³³

Figures 8 and 9 show that community membership is a strong predictor of vote choice, rivaling policy in terms of importance. Party affiliation does not appear to affect vote choice. Candidate gender has a marginally significant effect on vote choice; perhaps surprisingly given women’s underrepresentation in leadership positions (Empresa Peruana de Servicios Editoriales, 2019), community presidents prefer female candidates to male candidates, although the effect is weaker than for community membership.³⁴ Respondents also generally prefer candidates who campaign on water or education to those who promise electricity.³⁵

I argue above that reciprocity institutions should facilitate the election of coethnic candidates to subnational office. To test this, I examine whether preferences for community-member candidates—particularly those from different communities than the respondent—are strongest where reciprocity institutions are maintained. Table 3 repeats estimation of equation (2), this time subsetting to communities where reciprocity institutions are and are not present.

Table 3.

Effects of Candidate Community Membership on Respondent Vote Choice by Reciprocity Subgroups.

	Dependent variable
	Likelihood of supporting candidate
	(1)	(2)	(3)	(4)
Candidate from your community	0.237*** (0.039)	0.314** (0.145)	0.142*** (0.042)	0.408*** (0.114)
Candidate from another community	0.151*** (0.043)	0.193** (0.093)	0.114*** (0.038)	0.066 (0.116)
Constant	0.373*** (0.023)	0.333*** (0.054)	0.417*** (0.023)	0.366*** (0.069)
Evidence of reciprocity:
Self-reported	Yes	No	Yes	No
Lab-in-the-field	Yes	Yes	No	No
Observations	964	108	1,072	116
R²	0.040	0.065	0.015	0.113
Adjusted R²	0.038	0.047	0.013	0.097

Reference category is a candidate from the district capital. Standard errors clustered at the respondent level. Subgroups calculated based on 2012 Cenagro census responses and lab-in-the-field experiment. Behavioral evidence obtained from the lab-in-the-field results aggregated to the municipal level. Behavioral evidence exists when the municipal-level treatment effect is positive and significant (p < .1). Self-reported evidence obtained from the self-reported measure of ayni in the 2012 Cenagro census.

p< .1. **p< .05. ***p < .01.

Evidence on the existence of reciprocity institutions is collected from two sources. First, I draw on the self-reported, community-level measure of ayni taken from the 2012 census of Peruvian indigenous communities. In addition to this self-reported measure, I construct a second measure using the results of the trust game above, aggregated to the municipal level. Specifically, where the municipal-level average treatment effects are positive and significant ( $p < 0.1$ ), I code reciprocity institutions as being present. In addition to providing a second measure of the existence of reciprocity institutions, the behavioral indicator captures the existence of reciprocity institutions at the municipal level, where cross-community coordination is required to elect community members to local office. Conversely, the self-reported measure captures reciprocity institutions only within individual communities.

Table 3 suggests that the presence of reciprocity institutions (i.e., ayni) is associated with a stronger preference for community-member candidates, regardless of whether they are from the respondent’s community or another one. Where there exists no evidence for reciprocity institutions, there appears to be no special preference for candidates from another community. However, an absence of reciprocity institutions does not inhibit the emergence of in-group favoritism for candidates from the respondent’s own community.³⁶

The null finding on candidates from another community in Column 4 of Table 3 does not seem to be solely a function of low statistical power. First, as in the other columns, the finding for candidates from the respondent’s own community still emerges in Column 4, despite the relatively small number of observations. Second, the estimates in Column 2 are calculated using even fewer observations than those in Column 4, and in the former, the finding for candidates from another community remains strong and significant. Finally, in comparison to other columns, Column 4 presents a noticeable attenuation in the magnitude of the coefficient for candidates from another community, suggesting that the size of the effect—and not just the precision of the estimate—is reduced for this subgroup.

The results thus suggest that relative to other predictors, indigenous community membership is a particularly strong predictor of community leaders’ preferences over candidates for subnational elective office. Even candidates who belong to a different community than the respondent are preferred over those who do not belong to a community.³⁷ Yet, evidence for this preference is strongest where reciprocity institutions are present. Absent these institutions, indigenous community presidents do not seem to prefer candidates from communities other than their own, making cross-community coordination to elect coethnic candidates to subnational office more difficult.

The quantitative evidence presented above is consistent with qualitative insights I obtained from two field sites in Peru. The first is the municipality of Urubamba, which has 38 indigenous communities. Based solely on the number of communities, Urubamba seems an unlikely case for indigenous collective action.³⁸ Yet, given my argument, Urubamba is, in fact, a case in which we would expect indigenous groups to achieve greater political representation.

Communities in Urubamba were relatively late adopters of communal titles. The median year of communal title receipt was 1988; the median date of adoption for all Peruvian communities was 1980. Consistent with my theory, the long-standing absence of communal titles in Urubamba has corresponded to a marked persistence of reciprocity institutions; all but 3 of the 38 communities in the municipality report practicing ayni.

According to interviews with municipal officials and community presidents in Urubamba, communities have effectively leveraged these reciprocity institutions to achieve political representation. Despite the large number of communities, the presidents of these communities have generally coordinated to nominate only one community-member candidate per election cycle. To determine who will run in a given election as the representative of all communities, community presidents use traditional institutions of reciprocity: the 5 to 10 largest communities rotate in nominating one of their leaders to run. A president—and thus, the community he leads—will support the candidate from another community in the current election, knowing that a candidate from his community will be the chosen candidate in a future election. Absent reciprocity institutions, such coordination would be complicated as there would be no guarantee that the community nominating the current candidate would cede that ability in future elections.

Once nominated, the candidate who will represent all of Urubamba’s communities faces off against other candidates from the municipal capital, a town of about 3,000 people. If elected, the community-member candidate rewards her base of support. As a municipal official from Urubamba stated in an interview, “Communities are loyal. They have their [single] candidate and if that candidate is elected, he will give them everything, even if it means doing much less in the capital.”³⁹

Thus, in municipalities like Urubamba, where communal titles are a relatively recent innovation, traditional institutions of reciprocity have persisted. These institutions then facilitate cross-community coordination to elect indigenous leaders to subnational posts. Once elected, these candidates target distributive benefits to indigenous communities.

Paccha, a municipality in the department of Junín, has followed a very different pattern from Urubamba. Within the municipality, there are seven communities, which all obtained a title prior to 1957.⁴⁰ Consistent with my theory, none of these communities report preserving reciprocity institutions (Instituto Nacional de Estadística e Informática, 2014). As one community president told me, “There is too much concern with buying and selling. No one will do things in the old ways. No one here will work for free.”⁴¹

The absence of reciprocity institutions coincides with a failure by indigenous communities to coordinate around indigenous candidates for local office. In the 2018 municipal elections, 10 candidates ran for mayor of Paccha. Candidates from four of the seven communities ran in the elections, marking a strong contrast with Urubamba where—often—only one community-member candidate runs to represent all 37 communities. In the lead-up to Paccha’s 2018 election, one community president told me, “It will be hard for a community member to win. The votes aren’t there. . .There’s too much conflict among the communities here.”⁴² He was proven correct a year later when a candidate from the district capital won the mayorship.

Thus, both qualitative and quantitative evidence suggests that informally held communal land preserves reciprocity institutions, which serve to coordinate communities around shared indigenous candidates for subnational office. Communal titles, on the other hand, may undermine reciprocity institutions; as a result, indigenous groups may find it difficult to electorally coordinate their support for a single coethnic candidate.

Conclusion

In this paper, I highlight the important ways in which communal property can negatively affect the likelihood that indigenous groups achieve coethnic political representation. I first used historical data and a lab-in-the-field experiment to demonstrate that communal land titles appear to erode traditional reciprocity institutions while informal possession of communal land preserves them. I then provided evidence from a conjoint experiment, analysis of data scraped from candidate CVs, and qualitative fieldwork to show that the preservation of these institutions facilitates indigenous groups’ coordination around a single indigenous candidate for subnational office. Thus, while informal possession of communal land reinforces reciprocity institutions and thereby leads to long-term gains in coethnic political representation, communal titles have opposing effects.

The theory and evidence presented in this article make two distinct contributions. First, this article places certain limits on the expected welfare gains from communal property, particularly for indigenous groups. In this sense, formal titles serve as an impetus driving the transition from a moral economy to a market economy that scholars have long thought to be negative for indigenous peasants (Migdal, 1974; Paige, 1978; Scott, 1977, cf. Popkin, 1979). The observed effects of communal land titles should be compounded in contexts where private land titles are extended, and thus, land can be bought and sold.⁴³ Yet, the findings I present do not necessarily suggest that the aggregate effects of formal titles are negative. In fact, secure title may have positive welfare effects on domains outside of the ones I explore here. For example, the security of land access may increase productivity and income in indigenous communities.⁴⁴

Second, the evidence presented here outlines the conditions under which reciprocity institutions, a key mechanism underlying coethnic favoritism, persist, or fail (Habyarimana et al., 2009). Often, scholars describe the transformation, persistence, or erosion of reciprocity institutions as a—perhaps overly—functionalist response to market transformations (e.g., Polanyi, 1944; Thompson, 1971). My findings offer a way to understand when these institutions endure or wane, due to the presence or absence of communal land titles.

Ultimately, the dynamics around the extension of communal land titles play an important but largely understudied role in shaping indigenous groups’ relationship to the state. Existing research largely considers the economic and social effects of these institutions (Bottazzi & Rist, 2012; Cramb & Wills, 1990; Dippel, 2016; Sjaastad & Bromley, 1997). Yet, limited work examines how communal landholding shapes native populations’ access to political representation within the state. The research presented here suggests that communal titles may require indigenous groups to engage the state, but they may also—under certain conditions—reduce the ability of native communities to coordinate to achieve political voice. Future research may systematically investigate how other institutional features, such as electoral quotas, ethnic political parties, and reservations, ameliorate or exacerbate the negative representational effects of communal property observed in this paper.

Supplemental Material

sj-pdf-1-cps-10.1177_0010414021997157 – Supplemental material for The Representational Effects of Communal Property: Evidence from Peru’s Indigenous Groups

Supplemental material, sj-pdf-1-cps-10.1177_0010414021997157 for The Representational Effects of Communal Property: Evidence from Peru’s Indigenous Groups by Christopher L. Carter in Comparative Political Studies

Footnotes

Acknowledgements

For excellent comments, I thank the editors and two anonymous reviewers. I also thank my advisors--Thad Dunning, Andrew Little, Alison Post, and Ken Scheve--for thoughtful feedback and guidance throughout the project. I am grateful to Anna Callis, Anirvan Chowdhury, Fran Hagopian, Alisha Holland, and Steve Levitsky for their careful reading of and insightful comments on earlier versions of this paper. I am also grateful to seminar participants at Harvard University; University of California, Berkeley; the 2018 Program on Governance and Local Development Conference; and the 2018 annual meeting of the Midwest Political Science Association. Gabriella Wong and IPA Peru did a phenomenal job fielding the survey that contained the conjoint and lab-in-the-field. Replication code and data for this article available at . Research activities approved under Berkeley IRB (2016-09-9126). The experimental portions of this paper were pre-registered with EGAP (20170716AA).

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Christopher L. Carter

Supplemental Material

Supplemental material for this article is available online.

Notes

Author Biography

Christopher L. Carter is an Academy Scholar at the Harvard Academy for International and Area Studies and Research Associate at the Center on the Politics of Development at the University of California, Berkeley. He received his PhD in Political Science from the University of California, Berkeley in 2020.

References

Abramson

S. F.

Kocak

Magazinnik

(2019). What do we learn about voter preferences from conjoint experiments? Working paper.

Albó

(1991). El retorno del indio. Revista Andina, 9(2), 299–366.

Albó

(2008). Movimientos y Poder Indígena en Bolivia, Ecuador y Perú. CIPCA.

Austin

R. D.

(2009). Navajo courts and Navajo common law: A tradition of tribal self-governance. University of Minnesota Press.

Bansak

Hainmueller

Hopkins

D. J.

Yamamoto

(2020). Using conjoint experiments to analyze elections: The essential role of the Average Marginal Component Effect (AMCE). SSRN Paper, Rochester, NY.

Benton

A. L.

(2017). Configuring authority over electoral manipulation in electoral authoritarian regimes: Evidence from Mexico. Democratization, 24(3), 521–543.

Berg

Dickhaut

McCabe

(1995). Trust, reciprocity, and social history. Games and Economic Behavior, 10(1), 122–142.

Bottazzi

Rist

(2012). Changing land rights means changing society: The sociopolitical effects of agrarian reforms under the government of Evo Morales. Journal of Agrarian Change, 12(4), 528–551.

Brülhart

Usunier

(2012). Does the trust game measure trust? Economics Letters, 115(1), 20–23.

10.

Carter

C. L.

(2020). States of extraction: The emergence and effects of indigenous autonomy in the Americas (Doctoral dissertation). UC Berkeley.

11.

Carter

C. L.

(2021a). The autonomy-representation dilemma: Indigenous groups and distributive benefits in the Americas. Working Paper.

12.

Carter

C. L.

(2021b). Replication data for: The representational effects of communal property. Harvard Dataverse, V1. https://doi.org/10.7910/DVN/M4ZWL8

13.

Cramb

R. A.

Wills

I. R.

(1990). The role of traditional institutions in rural development: Community-based land tenure and government land policy in Sarawak, Malaysia. World Development, 18(3), 347–360.

14.

Davies

T. M.

(1974). Indian integration in Peru: A half century of experience, 1900-1948. University of Nebraska Press.

15.

Degregori

C. I.

(1998). Ethnicity and democratic governability in Latin America: Reflections from Two Central Andean Countries. In Agüero

Stark

(Eds.), Fault lines of democracy in post-transition Latin America (pp. 203–227). North-South Center Press/University of Miami.

16.

Díaz-Cayeros

Magaloni

Ruiz-Euler

(2014). Traditional governance, citizen engagement, and local public goods: Evidence from Mexico. World Development, 53, 80–93.

17.

Díaz-Polanco

(1998). La Autonomía, Demanda Central de los Pueblos Indígenas: Significado e Implicaciones. In Alta

Iturralde

López Bassola

(Eds.), Pueblos Indígenas y estado en América Latina (pp. 213–220.) Editorial Abya Yala.

18.

Dippel

(2016). Forced coexistence and economic development: Evidence from Native American reservations. In Anderson

T. L.

(Ed.), Unlocking the wealth of Indian nations (pp. 160–195). Lexington Books.

19.

Dubertret

Alden Wily

(2015). Percent of indigenous and community lands. Data file from Landmark: The Global Platform of Indigenous and Community Lands. www.landmarkmap.org

20.

Eisenstadt

T. A.

(2011). Politics, identity, and Mexico’s indigenous rights movements. Cambridge University Press.

21.

Empresa Peruana de Servicios Editoriales. (2019). Destacan Aprobación de Ley que Fortalece Rol de la Mujer en las Comunidades Campesinas. Andina.

22.

Estermann

(1998). Filosofía Andina: Estudio Intercultural de la Sabiduría Autóctona Andina. Abya-Yala.

23.

Field

(2007). Entitled to work: Urban property rights and labor supply in Peru. The Quarterly Journal of Economics, 122(4), 1561.

24.

Flórez

(2012). Rivalidades Comunales y Contiendas Electorales: Micropolítica en las Elecciones Distritales de Chuschi. In Hurtado

A. D.

(Ed.), Tensiones y transformaciones en comunidades campesinas (pp. 225–262). Pontificia Universidad Católica del Perú (CISEPA).

25.

Fonseca Martel

(1974a). La Minka y elWaje-Waje: Relaciones de Reciprocidad de una Región Andina del Perú. UNMSM.

26.

Fonseca Martel

(1974b). Modalidades de la Minka. In Alberti

Mayer

(Eds.), Reciprocidad e intercambio en los Andes peruanos (pp. 86–109). Instituto de Estudios Peruanos.

27.

Freire

Schwartz Orellana

S. D.

Zumaeta Aurazo

Costa

D. C.

Lundvall

J. M.

Viveros Mendoza

M. C.

Lucchetti

L. R.

Moreno Herrera

L. L.

Sousa

L. D. C.

(2015). Indigenous Latin America in the twenty-first century: The first decade (Technical report 98544). The World Bank.

28.

García

M. E.

(2005). Making indigenous citizens: Identities, education, and multicultural development in Peru. Stanford University Press.

29.

Guillet

(1980). Reciprocal labor and peripheral capitalism in the Central Andes. Ethnology, 19(2), 151–167.

30.

Habyarimana

Humphreys

Posner

D. N.

Weinstein

J. M.

(2009). Coethnicity: Diversity and the dilemmas of collective action. Russell Sage Foundation.

31.

Hainmueller

Hopkins

D. J.

(2015). The hidden American immigration consensus: A conjoint analysis of attitudes toward immigrants. American Journal of Political Science, 59(3), 529–548.

32.

Hainmueller

Hopkins

Yamamoto

(2014). Causal inference in conjoint analysis: Understanding multidimensional choices via stated preference experiments. Political Analysis, 22(1), 1–30.

33.

Hoffay

Rivas

(2016). The indigenous in Latin America: 45 Million with little voice. Global Americans. Global Americans, September 1, 2016. https://theglobalamericans.org/2016/08/indigenous-latin-america-45-millionlittle-voice/

34.

Htun

(2016). Inclusion without representation in Latin America: Gender quotas and ethnic reservations. Cambridge University Press.

35.

Huenchulaf Cayuqueo

(1998). La Temática Indígena en el Concierto Internacional. In Alta

Iturralde

D. A.

López-Bassols

M. A.

, (Eds.), Pueblos Indígenas y Estado en América Latina (pp. 147–159). Editorial Abya-Yala.

36.

Hurtado

A. D.

(2012). Tensiones y Transformaciones en Comunidades Campesinas. Pontificia Universidad Católica del Perú, CISEPA.

37.

Instituto Nacional de Estadística e Informática. (2011). Encuesta Nacional de Hogares. Lima, Peru. https://webinei.inei.gob.pe/anda_inei/index.php/catalog/369

38.

Instituto Nacional de Estadística e Informática. (2014). IV Censo Nacional Agropecuario. Lima, Peru. https://proyectos.inei.gob.pe/CenagroWeb/

39.

Jackson

J. E.

Warren

K. B.

(2005). Indigenous movements in Latin America, 1992–2004: Controversies, ironies, new directions. The Annual Review of Anthropology, 34(1), 549–573.

40.

Keefer

Vlaicu

(2008). Democracy, credibility, and clientelism. Journal of Law, Economics, & Organization, 24(2), 371–406.

41.

Lanjouw

J. O.

Levy

P. I.

(2002). Untitled: A study of formal and informal property rights in urban Ecuador*. The Economic Journal, 112(482), 986–1019.

42.

Long

Roberts

B. R.

(2014). Peasant cooperation and capitalist expansion in Central Peru. University of Texas Press.

43.

Madrid

R. L.

(2012). The rise of ethnic politics in Latin America. Cambridge University Press.

44.

Mallon

F. E.

(2014). The defense of community in Peru’s Central Highlands: Peasant struggle and capitalist transition, 1860-1940. Princeton University Press.

45.

Mannheim

(2013). The language of the Inka since the European invasion. University of Texas Press.

46.

Mayer

(2018). The articulated peasant: Household economies in the Andes. Routledge.

47.

Mendoza

R. R.

(2002). Legislación Peruana sobre Comunidades Campesinas. UNMSM.

48.

Migdal

J. S.

(1974). Peasants, politics and revolution: Pressures toward political and social change in the third world. Princeton University Press.

49.

Ministerio de Vivienda, Construccion, y Saneamiento. (2009). Directorio de Comunidades Campesinas del Peru 2009.

50.

Nuñez Palomino

P. G.

(1996). Derecho y Comunidades Campesinas en el Perú, 1969-1988. Centro Bartolomé de las Casas.

51.

Olson

(1971). The logic of collective action. Harvard University Press.

52.

Paige

J. M.

(1978). Agrarian revolution. Simon and Schuster.

53.

Place

Hazell

(1993). Productivity effects of indigenous land tenure systems in Sub-Saharan Africa. American Journal of Agricultural Economics, 75(1), 10–19.

54.

Polanyi

(1944). The great transformation: The political and economic origins of our time. Beacon Press.

55.

Popkin

S. L.

(1979). The rational peasant: The political economy of rural society in Vietnam. University of California Press.

56.

Postero

N. G.

Zamosc

(2004). The struggle for indigenous rights in Latin America. Sussex Academic Press.

57.

Remy

M. I.

(2013). Historia de las Comunidades Indígenas y Campesinas del Perú. IEP.

58.

Rice

Van Cott

D. L.

(2016). The emergence and performance of indigenous peoples’ parties in South America: A subnational statistical analysis. Comparative Political Studies, 39(6), 709–732.

59.

Rights and Resources Initiative. (2015). Who owns the world’s land? A global baseline of formally recognized indigenous community land rights. Washington, DC: Rights and Resources Initiative.

60.

Scott

J. C.

(1977). The moral economy of the peasant: Rebellion and subsistence in Southeast Asia. Yale University Press.

61.

Seligmann

L. J.

(1995). Between reform & revolution: Political struggles in the Peruvian Andes, 1969-1991. Stanford University Press.

62.

Sjaastad

Bromley

D. W.

(1997). Indigenous land rights in Sub-Saharan Africa: Appropriation, security and investment demand. World Development, 25(4), 549–562.

63.

Sobel

(2005). Interdependent preferences and reciprocity. Journal of Economic Literature, 43(2), 392–436.

64.

Stavenhagen

(2000). Towards the right to autonomy in Mexico. In Cal

A. B.

Mayor

(Ed.), Indigenous autonomy in Mexico (pp. 10–23). IWGIA.

65.

Stern

S. J.

(1993). Peru’s Indian peoples and the challenge of Spanish conquest: Huamanga to 1640. University of Wisconsin Press.

66.

Thompson

E. P.

(1971). The moral economy of the English crowd in the eighteenth century. Past Present, 50(1), 76–136.

67.

Van Cott

D. L.

(2001). Explaining ethnic autonomy regimes in Latin America. Studies in Comparative International Development, 35(4), 30–58.

68.

Van Cott

D. L.

(2005). Building inclusive democracies: Indigenous peoples and ethnic minorities in Latin America. Democratization, 12(5), 820–837.

69.

Van Cott

D. L.

(2009). Radical democracy in the Andes. Cambridge University Press.

70.

Webb

R. C.

(2013). Conexión y Despegue Rural. Universidad San Martin de Porres. Fondo Editorial.

71.

Winder

(2014). The impact of the comunidad on local development in the Mantaro Valley. In Long

Roberts

B. R.

(Eds.), Peasant cooperation and capitalist expansion in Central Peru (pp. 209–240). University of Texas Press.

72.

Wutich

Beresford

Carvajal

(2017). Ayni real and imagined: Reciprocity, indigenous institutions, and development discourses in contemporary Bolivia. The Journal of Latin American and Caribbean Anthropology, 22(3), 475–494.

73.

Yashar

D. J.

(1998). Contesting citizenship: Indigenous movements and democracy in Latin America. Comparative Politics, 31(1), 23–42.

74.

Yashar

D. J.

(2005). Contesting citizenship in Latin America: The rise of indigenous movements and the postliberal challenge. Cambridge University Press.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.92 MB