Itaú BBA - Setting the stage for a pause

Scenario Review - Mexico

< Back

Setting the stage for a pause

March 9, 2018

We do not expect further interest rate increases.

Please see the attached file for all graphs.

The tides are moving in favor of the anti-establishment candidate, Andrés Manuel López Obrador (AMLO), who widened his lead in February (according to polls) and is showing a lower rejection rate than his rivals. 

Recent trade tariffs imposed by the U.S. excluded Mexico and Canada. While risks are significant, we still think a NAFTA deal is likely in 1H18.  

We have revised our GDP growth forecasts downward for 2018 (to 1.8%, from 2.1%) and 2019 (to 2%, from 2.4%).

We see the recent communication from the central bank as consistent with our view that a pause in April is likely. The board, or at least most of its members, seems less willing to be as reactive to the Fed and the Mexican peso as before, focusing instead on the future path for inflation.

AMLO is getting stronger

The tides are moving in favor of the anti-establishment candidate, Andrés Manuel López Obrador (AMLO), who widened his lead in February (according to polls). The average of the four most reliable polls (Mitofsky, Buendía Laredo, El Financiero, and Reforma) indicated that AMLO widened his lead over the runner-up, Anaya, by 2.5 pp (from 4.6% in January to 7.1% in February)[1]. Specifically, the results were as follows: AMLO (29.7% in February, from 27.8% in January), Anaya (22.6%, from 23.2%), Meade (16.2%, from 17.1%), and Zavala (4.7%, from 4.4%).

Moreover, two recent developments cast doubt on whether “tactical voting” (which played against AMLO in the past) will materialize with similar strength as in previous elections: the decrease of AMLO’s rejection rate, and the heightened confrontation between the ruling party (PRI) and the PAN/PRD alliance. The polls that publish rejection rates (Reforma, El Financiero, Buendía Laredo) show a falling trend for AMLO (26.5% in February, from 33% in October) and an increase for Meade (39%, from 13%) and, to lesser extent, for Anaya (30%, from 25%). So the polls published in February 2018 are the first ones to show a lower rejection rate for AMLO relative to his competitors. In the PRI’s camp, it is proving hard for Meade to decouple from the high PRI’s rejection rate (which stands around 50% according to most polls). Also, we note that the heightened confrontation between the PRI and PAN/PRD is risking the images of both political forces, while AMLO remains on the sidelines. 

NAFTA 7th round: tarnished by threat of global protectionism

The 7th round of NAFTA renegotiations – held in Mexico City between February 25 and March 5 – continued finalizing deals on additional chapters and breaking the ice on the more controversial issues, but it was tarnished by news that the U.S. will impose tariffs on steel and aluminum on imports from the rest of the world. These tariffs, unlike the U.S. safeguards on solar panels and washing machines announced in January, would be implemented under the “national security exception,” which is codified not only in the U.S. Trade Law but also in NAFTA and the WTO. We note the “national security exception” is rarely used (the last time the U.S. imposed tariffs by invoking it was in the Reagan administration) and could lead to retaliation from big economies such as China and the EU. Nevertheless, President Trump granted an exemption to Mexico and Canada which, he argued, could be made permanent or revoked depending on the outcome of the NAFTA renegotiation. In any case, the NAFTA 7th round finalized the deal on regulatory practices, sanitary & phytosanitary measures, and transparency (6 of 30 negotiation points have now been completed) and continued the back and forth on the controversial issues (mainly dispute settlement and rules of origin). We highlight that the tone of negotiators from the three sides, at the closing press conference, was constructive. 

Our base case is that a NAFTA deal will be reached during the first half of this year. The Mexican Senate will enter a recess in May, but special commissions will continue to operate and could ratify a renegotiated NAFTA agreement anytime until the last day of August 2018 (before the new Congress starts on September 1). However, the fact that talks will likely overlap with Mexico’s presidential campaign season and the new trade measures announced by the Trump administration are risks (at least for timely completion of a deal), even though the AMLO camp’s tone on NAFTA has been more positive.

Moderate growth ahead, amid elections and NAFTA uncertainty

Mexico’s GDP growth slowed down in 2017 – in the midst of uncertainty (with NAFTA renegotiation and the forthcoming presidential elections standing out as key risks), plunging oil output and the doubling of inflation (which ate through real wages). The monthly GDP proxy (IGAE) grew 1.1% year over year in December, and GDP growth for 4Q17 posted 1.5% (below the flash estimate published by INEGI last month). According to calendar-adjusted data reported by the Statistics institute (INEGI), GDP expanded 1.5% year over year in 4Q17 (from 1.6% in 3Q17). Looking at the full-year figures, we note that in 2017 GDP growth slowed to 2% (from 2.9% in 2016), with slower expansion across all the main sectors: Services (3%, from 3.9%), with retail sales weakening substantially; Industrial production (-0.6%, from 0.4%), with manufacturing outperforming mining and construction; and the small and volatile primary sectors (3.3%, from 3.8%), which basically represent agriculture.

At the margin, GDP recovered momentum, as expected, following a weak 3Q17 when the country was hit by natural disasters. In December, the seasonally-adjusted monthly GDP proxy gained 0.7% from the previous month and quarter-over-quarter annualized growth jumped to 3.2% (from -0.7% qoq/saar in 3Q17). Importantly, this rebound is not only about the normalization of oil output (which bore the brunt of the hurricanes and earthquakes in 3Q17) but also visible in GDP growth excluding mining and the volatile primary sectors (3.4% qoq/saar in 4Q17, from 0.1% in 3Q17). Granted, the service sector, particularly retail, also took a hit from the natural disasters.

We have revised our GDP growth forecasts downward for 2018 (to 1.8%, from 2.1%) and 2019 (to 2%, from 2.4%). The factors playing against economic growth in the short-term are tight macro policies (fiscal and monetary) and the uncertainties associated with NAFTA and elections (which put investment decisions on hold). On the plus side, we note that the fiscal drag will be smaller in 2018 relative to 2017. Moreover, a stronger U.S. economy will likely stimulate Mexico’s manufacturing exports. Finally, we see lower inflation improving the growth of real wages. 

Inflation is trending down

Inflation kept falling in February. Headline inflation decreased to 5.34% year over year in February (from 5.55% in January), while core inflation decreased to 4.27% (from 4.56%) during the same period. Moreover, the diffusion index and measures of inflation at the margin decreased meaningfully. The diffusion index, which tracks the percentage of items in the CPI basket with inflation higher or equal to four, went down to 62.5% (from 69.4% in January), reaching the lowest level in a full year. Also, seasonally-adjusted three-month annualized inflation decreased to 3.41% (from 5.22% in January) for the CPI and to 3.74% (from 4.46% in January) for the core index.

We expect inflation to reach 3.7% by the end of 2018 (below median market expectations of 4.1%, according to the central bank’s last survey). The more benign evolution of the currency will be the key driver, as the backlog of exchange-rate depreciation (60% between 2014 and 2016) has probably died out and pass-through is now actually exerting downward pressure. Moreover, we see further room for the normalization of non-core inflation. Inflation for regulated/administered items is also relevant for the currency, and non-core food inflation is standing at a very high level (9.7% vs. 10-year median of 5.6%). 

Banxico puts more emphasis on inflation-forecast targeting

The Central Bank of Mexico (Banxico) hiked the reference rate by 25 bps (to 7.50%) at its first meeting of the year, in line with our call and an almost unanimous market consensus. Given risks related to NAFTA, elections and monetary policy in the U.S. (amid still high inflation readings), the central bank kept the doors open for further hikes by saying that “monetary policy will act, if necessary, firmly and opportunely to ensure inflation expectation anchoring and the convergence of inflation to the target.” Furthermore, the board’s view on the balance of risks for inflation is unchanged (and continues tilted to the upside). 

However, when mentioning the factors it will monitor for the upcoming decisions, the central bank did not prioritize the relative monetary policy stance between Mexico and the U.S. (moving it to second position). Furthermore, the board affirmed that the February hike already took into account the expected rate increase by the Fed for March. Board Member Javier Guzman, considered to be in the hawkish camp, reaffirmed the same message.

Additionally, the central bank published the first quarterly inflation report of the year, with a greater emphasis on the role that inflation forecasts will play in the board’s decision framework. The bank now publishes the values of average annual inflation (headline and core) that it expects for the next eight quarters, comparing them with those in the previous report. During his presentation, Governor Díaz de León argued that deviations from these forecasts will be important to determine future adjustments in monetary policy. 

In all, we see the recent communication as consistent with our view that a pause in April is likely. The board, or at least most of its members, seems less willing to be as reactive to the Fed and the Mexican peso as before, focusing instead on the future path for inflation. Although risks are significant, we expect Mexico’s next rate move to be a cut (in the second half of the year). 

Lower current account deficit

The current account deficit (CAD) narrowed by 0.5% of GDP in 2017 (to an easily financed 1.6% of GDP). The CAD narrowed on a record-high non-energy trade surplus and solid transfers (driven by the pick-up of the US economy) and a smaller net income deficit (reflecting lower profit remittances from foreign firms operating in Mexico). We have revised our current account deficit forecast for 2018 (to 1.3% of GDP, from 1.6% of GDP) and 2019 (1.4% of GDP, from 1.7% of GDP). The CAD will likely narrow a bit more, as manufacturing exports continue accelerating, while internal demand expands at a more moderate pace. 

Monitor the risks

Mexico’s economic outlook is subject to an unusually high degree of uncertainty. Besides the risks related to the trade relations with the U.S., market participants fear a change in the economic policy framework after the elections. If in the near term it becomes clear that economic policies will remain untouched and a NAFTA agreement is reached, the macro outlook would be more benign than we currently forecast (with higher growth, stronger currency, lower inflation and deeper rate cuts). On the other hand, if Mexico moves to unorthodox policies while there is no agreement on NAFTA, then the scenario becomes more adverse than we currently forecast. We note that the two sources of risk that Mexico is facing (NAFTA and elections) are not independent. Clearly, the harsh rhetoric of the U.S. government aimed at Mexico fuels a more nationalistic campaign in the country. At the same time, a change in government could derail (or postpone) talks on NAFTA. Finally, it is important to note that Mexico seems more prepared to deal with shocks than in the recent past, given the substantial narrowing of the twin deficits and the tight monetary policy (turning short positions on the Mexican peso more expensive).


[1] It is worth mentioning that El Financiero presents the results adding up to 100%, without counting the percentage of “no answers” (published in a footnote). So we adjust these results by multiplying them by the factor “1 minus the percentage of no answers” in order to make them comparable with other polls.


João Pedro Bumachar
Alexander Muller

< Back