Introduction Excerpt for Indicators of Democracy

Indicators of Democracy
The Politics and Promise of Evaluation Expertise in Mexico
Diana Graizbord

INTRODUCTION

IN JANUARY 2013, JUST WEEKS after assuming office, Mexico’s president Enrique Peña Nieto introduced La Cruzada Contra el Hambre (The Crusade Against Hunger), his flagship anti-poverty strategy, in Las Margaritas, Chiapas.1 Speaking to a large and cheering crowd of supporters, Peña Nieto promised that under his administration, “hunger” would be eradicated, and 7.4 million “extremely poor” people across four hundred municipalities would be lifted out of poverty. Las Margaritas was the first stop on a much-anticipated national tour during which Peña Nieto would visit many poor, largely Indigenous municipalities to promote the Cruzada directly to its potential beneficiaries.

The event was, in many ways, unremarkable. Promotional tours during which incoming or sitting presidents make direct promises to and pose for pictures with cheering crowds of supporters have long been a routine part of the political theater that characterizes Mexico’s strong brand of patrimonialism and highly presidential system.2 Nevertheless, President Peña Nieto’s announcement that day caused a media stir. As it turned out, although there had been talk of such a program since the presidential election in July 2012, the event in Chiapas was the first time that the president had provided any specific details about the Cruzada, and the details he provided fell short of what many had hoped to hear. In op-eds published in Mexico’s major newspapers, political blogs, and radio and television interviews, political opponents, journalists, social policy experts, and civil society leaders criticized the president’s lack of transparency and demanded that his administration justify various technicalities related to the design and targeting of his new anti-poverty strategy.

Of particular concern to these critics was what they saw as the program’s questionable targeting scheme. Some commentators revealed surprise at the muddied use of key concepts such as “extreme poverty” and “hunger” and noted that hunger was not an indicator for which the Mexican state produced official metrics. With no way to determine the population affected by hunger, the program’s targeting would necessarily be arbitrary or, worse, politically motivated. Indeed, many were convinced that the program’s targeting was driven by political patronage. These commentators speculated that information about the names and locations of the four hundred municipalities selected to receive the program would prove that the Cruzada was being deployed in return for the support shown by the residents of these municipalities for the president’s political party, the Partido Revolucionario Institucional (Institutional Revolutionary Party; PRI), in the recent presidential election. Others, also convinced of the political motivations driving the program’s design, accused the president’s team of targeting certain municipalities not as a payment for votes already cast, but the opposite. According to them, the Cruzada was an attempt to secure new voters before the next election in municipalities that had previously supported opponents of the PRI.

With Peña Nieto’s team providing little information about the logic and decision making behind the Cruzada, some policy experts and data journalists began to reverse-engineer the program. Using data from Mexico’s national statistical agency and the latest official poverty estimates produced and published by Mexico’s Consejo Nacional de Evaluación de la Política de Desarrollo Social (National Council for the Evaluation of Social Development Policy; CONEVAL)—the federal agency responsible for defining poverty indicators and evaluating social policy—they attempted to establish how, from the 11.4 million people whom CONEVAL officially classified as “extremely poor,” the president’s team had arrived at the 7.4 million figure. Using the same data and working backward from Mexico’s close to 2,500 municipalities, they attempted to estimate why only 400 municipalities had been selected to receive support. The public policy think tank México Evalúa, for example, published a report based on a statistical analysis its researchers had conducted that showed that among municipalities classified as “poor” by CONEVAL, those with local elections scheduled for the coming year had a much higher probability of being selected for the program than those that did not (México Evalúa 2013). On the digital news site Animal Político, a political commentator and data scientist published a blog examining the same data that México Evalúa had used along with a series of published reports and methodological documents produced by CONEVAL in an attempt to establish which of Mexico’s three official poverty lines and related benchmarking indicators the Cruzada’s designers might have considered. In it, he reported that the program had not properly made use of any of Mexico’s official poverty metrics, concluding that even if the program’s targeting was not driven by electoral interests, it lacked an evidence base (Merino 2013).

Eventually, the legislature took up the matter, calling the head of the Secretaría de Desarrollo Social (Ministry of Social Development; SEDESOL) Rosario Robles and top social development administrators to a special session of the Cámara de Diputados (Chamber of Deputies). In a public session lasting several hours, the president’s team defended the Cruzada against claims of patronage politics and poor design, claiming that the program was based on evidence and that, despite claims to the contrary, the administration’s technical team had indeed used CONEVAL’s data and metrics in their program design.3 “We have based all of our analysis on the information obtained from CONEVAL,” Robles said. In front of a poster board on which was printed a simple Venn diagram, she continued, “The strategy is directed toward the population in extreme poverty, which suffers ‘nutritional deprivation.’ According to CONEVAL’s data, there are 28 million people with ‘nutritional deprivation’ and 11.7 [million] in ‘extreme poverty.’” Pointing to the center of the diagram, Robles stated, “The intersection between these two phenomena in the center indicates 7.4 million extremely poor people with nutritional deprivation across the territory. This is our target population.” When the president spoke of “hunger,” she continued, he had meant “nutritional deprivation,” suddenly invoking the official indicator defined by CONEVAL and calculated using the agency’s official poverty measurement method. As legislators pored over the program’s targeting scheme and compared it to CONEVAL’s official data, they peppered Robles and the national directors of the Cruzada with questions. Despite officials’ insistence that they had used CONEVAL’s metrics and data to inform the design and determine the targeting scheme of the program, legislators remained unconvinced. With the issue of what “hunger” meant and whether the targeting scheme could be trusted unresolved, the special legislative committee instructed CONEVAL to monitor and evaluate the Cruzada throughout the remainder of Peña Nieto’s term.

As these scenes were playing out in the national media and the legislative chambers, in CONEVAL’s headquarters—located just off Mexico’s busy Periférico ring road in a four-story gray stucco building—a different set of scenes unfolded. For many CONEVAL staffers, the months that followed Peña Nieto’s victory in the 2012 presidential election had been filled with anxiety and speculation about whether their agency would survive the political transition. When Peña Nieto took office and began to name members of his transition team and other loyalists to cabinet positions, anxiety grew. Some of the administrative and technical staff in the evaluation division worried that Gonzalo Hernández Licona, the affable Oxford-trained economist who had served as the executive secretary since the agency’s founding in 2005, would be replaced. The more cautiously optimistic staff interpreted the silence from Peña Nieto’s administration as evidence of CONEVAL’s staying power, reminding naysayers that the agency had already survived one presidential transition—although even they were certain that the agency would face severe budget cuts. Luisa, a young data analyst who had been working at CONEVAL for just over two years, asked me over coffee one afternoon if I thought that it was too late for her to apply to graduate school in the United States. She knew some deadlines had passed and others were fast approaching but was considering giving it a shot since she was concerned that soon she would be left without a job. “A ver si no nos cierran el changarro,” she said, worried that the new administration would force CONEVAL to “close up shop.”

During these months, in an effort to safeguard itself from possible elimination, CONEVAL began hosting a series of high-profile public-facing events commemorating their work and showcasing the benefits of monitoring and evaluation (M&E) and its prospective role for the incoming administration. At one event, held at the Camino Real Hotel in Mexico City’s swanky Polanco neighborhood, I joined civil servants, policy managers, international development practitioners, consultants, social scientists, and members of the press for three days of seminars on experimental methods, performance-based budgeting models, poverty measurement and identification techniques, and workshops on using statistical software packages and understanding the differences between diagnostic instruments and process evaluations. In addition, each day a series of panels featured Mexican technocrats and experts relaying the challenges and achievements that had marked the institutionalization and everyday implementation of M&E in Mexico in previous years. During afternoon media events, awards were given to federal social program officers and representatives of state and local governments for their effective use of evaluation. The atmosphere was mostly celebratory, though hints of existential anxiety occasionally peeked through. In a well-attended plenary on the last day, Gonzalo Hernández Licona spoke about the history of M&E in Mexico and CONEVAL’s work to encourage its use throughout the federal bureaucracy. When a member of the audience asked about future challenges facing the agency, he paused briefly before saying, “We’re fighting a political battle here in Mexico. Evaluation is hard work, but ultimately, we’re working to create and consolidate democracy day in and day out!”



 

Notes

1. For a detailed account of the Cruzada and how it fits within Mexico’s social policy landscape, see Cejudo and Michel (2015, 2017); Yaschine, Ochoa, and Hernández (2014).

2. The Mexican writer and public intellectual Octavio Paz, for example, long characterized the generalized veneration expressed by Mexican society toward the president, who in turn has historically performed a strong paternalistic role. The history of Mexican presidentialism is vast; for classic treatments, see Cosío Villegas (1974) and Meyer (2000). For a review of the literature, see Weldon (1997).

3. I watched the legislative session on the Cruzada Contra el Hambre in which the Rosario Robles participated live on YouTube, where it was broadcast on the Cámara de Diputados channel. The quotes I include in this paragraph are from the notes I recorded while viewing the session.

Back to Excerpts + more