Presidential Approval and the Recall Referendum in Mexico

El Financiero’s national public opinion polls on the matter offer various indicators of popular support for President López Obrador, notably including his approval ratings and referendum voting preferences. The latter fell to 54 percent according to the most recent poll conducted in February. It is a majority level of public support, yes, but it is also the lowest record the President has obtained since he took office Disapproval soared to 43 percent only weeks before the referendum, mainly because of news that his oldest son had lived in a luxury house in Houston.

Meanwhile, preferences on the Revocación de Mandato were measured with two questions. For many months, El Financiero asked respondents, “If the referendum on the Revocation of President López Obrador's Mandate were held today, how would you vote?” It then supplemented that question with the official wording that will appear in the referendum ballots on April 10th: “Do you agree that the mandate of Andrés Manuel López Obrador, President of the United Mexican States, should be revoked due to a loss of confidence or that he stay in the Republic’s Presidency until the end of his term?” 

The El Financiero question shows that 52 percent support AMLO to continue in office; 42 percent prefer that he does not continue. With minor variations, these are the same numbers from the presidential approval question. 

However, the official wording yields a 63 percent support for AMLO to "remain in office until his term ends," and 30 percent prefer his mandate to be "revoked due to a loss of confidence." This is a more favorable outcome for the President. The official wording gives 11 extra points in the President's favor, plus or minus the margin of error. 

Having both referendum questions in the poll allows us to see that one is a closer reflection of presidential approval, while the other – the official question – is not. 

If the recall referendum is supposed to be an institutional mechanism to channel and express citizen dissatisfaction with the current government in general, or with the President in particular, the official wording does not fully comply with that purpose.

The relationship between approving the president and preferring him to remain in office is very strong with the simple wording from El Financiero: 92 percent of those who strongly disapprove of the President prefer that he does not continue in office. The most intense discontent translates into a near-unanimous voting intention for repeal.

But that relationship weakens when the official wording is used: of those who strongly disapprove of the President's job, 72 percent would vote for a recall, and 28 percent prefer he remains in office. Only among those who disapprove strongly of AMLO, there is a 20-point difference between one wording and another. The polls also show a 20-point difference among those who approve somewhat of his job: 75 percent would vote for repeal in the simple wording, and 55 percent with the official wording.  

The official question produces an effect of continuity even among the most dissatisfied citizens. One possible interpretation is that the phrase "let him remain in office until his term ends" is an institutionally acceptable response, which dampens anger. It may even reflect the custom that presidents are elected for six years, even if they are disliked.

If the dice are loaded in the official question, the bias favors continuity, not change. This may help President López Obrador not only to have a fresh and renewed legitimacy on April 10, but also for him to have a more favorable outcome than what the distribution of opinions should weigh. 

The fact that the opposition has called on constituents to abstain from voting in the referendum may also help the President to have a more favorable result. That is strange move because the mobilization of discontent has been a successful electoral strategy in the last decade around the world, including that of AMLO in 2018. 

By demobilizing discontent, the opposition lowers the volume of the dissatisfied voices of the citizenry, potentially contributing to the fact that the referendum does not reflect its proportional weight in the results: 43 percent, if disapproval in February is considered, versus 30 percent, if support for revocation is considered in the official question. 

Any result lower than those in the recall option on April 10 may be attributable to the demobilization of discontent by the opposition. The challenge for Morena, the president’s party, is to mobilize as much support as possible, which was until recently complicated by legal provisions that restricted such mobilization efforts on behalf of legislators. El Financiero's national poll estimates a likely turnout of 18 percent and, possibly, of up to 27 percent. The predictions are rather high when compared to the 7 percent turnout for a referendum last year, but they are certainly much lower than the 40 percent required for the referendum to be legally binding. 

Given the extra help from both the official question and from the demobilization of discontent by the opposition, the outcome of the recall referendum is likely to reflect a different distribution of opinions to what Mexican society currently holds. We will see how voters actually behave. 

[*]An earlier version of this was published in Spanish as "Aprobación y revocación," El Financiero, March 4th, 2022.