Itaú BBA - Domestic and external uncertainties return

Scenario Review - Mexico

< Back

Domestic and external uncertainties return

November 6, 2017

We have reduced our growth forecast for this year, but revised our inflation forecast up.

Please see the attached file for all graphs.

• The latest round of NAFTA renegotiation raised the odds of a break-up as some of the U.S. demands were rejected by Mexico and Canada. 

In addition, the right-wing PAN party has split, benefiting the anti-establishment candidate Andrés Manuel Lopez-Obrador (AMLO), but also the ruling PRI.

Mexico’s FX commission, formed by the Central Bank and the Ministry of Finance, stepped up its FX swap interventions. Rate hikes seem unlikely for now, but the prospects for monetary easing receded further.

We have revised our GDP growth forecast for 2017 (to 2.1%, from 2.3%) in response to the slowdown observed in 3Q17 (largely explained by the one-off effects the earthquakes). Moreover, we now expect annual CPI inflation at 5.9% by the end of 2017 (from 5.7% in our previous scenario) because inflation for core goods (tradables) is falling slower than expected.

NAFTA uncertainty intensifies

The fourth round of NAFTA renegotiation, held on October 11-17 in the U.S., raised the odds of a NAFTA break-up as some of the U.S. demands were rejected by Mexico and Canada. Still, the three member countries announced that they will continue negotiating into 2018 (when both Mexico and the U.S. will hold elections). The initial schedule laid out in August (seven to nine rounds, concluding by December 2017) was modified, with negotiations now expected to end in 1Q18. There are mainly four U.S. demands that could lead to deadlock: i) stricter rules of origin in the auto sector (increase in regional content requirement, to 85% from 62.5%, and inclusion of a previously non-existent minimum U.S. content requirement of 50%); ii) the elimination of Chapter 19, which governs trade remedies (banning safeguards and limiting the use of anti-dumping and countervailing duties more effectively than the WTO); iii) a “sunset clause,” whereby NAFTA would be renegotiated every five years or, otherwise, be terminated; and iv) seasonal windows in agriculture, which means that tariff-free access to the U.S. market for specific agricultural goods (tomatoes and strawberries, among others) would be restricted to the time of the year when there is no local production/harvest.

In spite of higher risks, we note that the U.S. has incentives to strike a deal. By withdrawing from NAFTA, President Trump would be putting at risk the approval of his tax reform – arguably a more important objective than NAFTA – by engaging in confrontation with Republican Senators from agricultural constituencies (whose support will be critical to pass the tax reform). From an economic perspective, if NAFTA ceases to exist, U.S. exporters face the risk of being penalized with much higher tariffs (vis-à-vis Mexican exporters), considering that the rates at which the U.S. and Mexico bound their tariffs under the WTO framework are, on average, 3.5% and 36.2% , respectively. 

With no legal precedent (for quitting free trade agreements) to refer to, it is not clear whether President Trump can withdraw the U.S. from NAFTA without the approval of Congress. Granted, Article 2205 of NAFTA states that any member country can withdraw unilaterally, announcing its intention six months in advance, and President Trump could invoke this article through an Executive Order (as was rumored in April). However, aspects of U.S. legislation seem to indicate that Congress would have a final say in the matter. According to the Constitution, Congress has the power to regulate commerce with foreign nations (the Executive Branch has authority over managing international relations, among other responsibilities). In the past, the Supreme Court has ruled that that the President is responsible for negotiating “treaties” (which require a 2/3 majority approval vote in the Senate). The point is that NAFTA is not technically a “treaty”, but rather a “congressional-executive agreement” approved under the framework of the fast-track mechanism, whereby the President submits a trade-agreement-implementing bill to Congress (only requiring a simple majority vote in both houses for approval). Therefore, pulling out from NAFTA would ultimately demand the repeal of the NAFTA Implementation Act, and – according to the Constitution – only Congress has the power to repeal laws. Supporters of the view that Trump can withdraw without Congress’s approval, however, argue that most of the NAFTA Implementation Act’s provisions would be repealed automatically once the president decides to abandon the trade agreement.

Also, some concessions to the U.S. from Mexico are feasible, even in the case of the most controversial issues (listed above). Middle ground can be reached on rules of origin. Recent news revealed that the Mexican government is working on a counterproposal for the auto sector (aimed at increasing regional content requirements, in order to substitute imports from outside of North America) to be presented at the fifth NAFTA renegotiation round, scheduled for November 17. Regarding the elimination of Chapter 19, arguably the thorniest issue (because the Canadian government and the Mexican Senate are fiercely opposed to it), a possible compromise could be the legitimization of safeguards (banned under NAFTA, but allowed by WTO rules); this would provide the U.S. government with more discretion to protect its domestic industries and, at the same time, avoid violating multilateral trade legislation. Overall, as specified in the “Summary of Objectives for the NAFTA Renegotiation” published by the U.S. government in July, the priority is to reduce its trade deficit. On the Mexican side, the head negotiator, Ildefonso Guajardo, has repeatedly stated that Mexico is willing to help the U.S. improve its trade balance through trade expansion between the two countries rather than trade restrictions. This can be done through the liberalization of services trade, where the U.S. is more competitive and actually has a trade surplus versus Mexico (barriers prevail in digital trade and cross-border data flows, among other areas), coupled with stronger protection for intellectual property rights (to boost sales of IP-intensive goods and services in Mexico). Adherence to tougher labor standards, like those to which Mexico already agreed in the (now-defunct) Trans-Pacific Partnership agreement, might also help to sugarcoat the deal for the U.S.

The economic impact of losing NAFTA for Mexico would be small in terms of foreign trade flows – assuming that trade migrates to the framework of the WTO – but could increase through falling investment. In a previous report (See “Reshaping NAFTA,” Resende and Müller, August 2017), we estimated that the shift of NAFTA to WTO’s most-favored-nation average tariffs (not “bound” tariffs, the highest level allowed without infringing WTO rules) would only wipe out USD 6.5 billion of Mexico’s trade (exports + imports) to the U.S. and Canada (that is, 1.3% of the annual trade exchange with these two countries). On the investment side, we acknowledge that investment flows into Mexico’s tradable sectors (mostly manufacturing) that were attracted by NAFTA in the past could diminish; but at the same time, we emphasize that there does not seem to be a very large difference between NAFTA and the investment protection framework for ex-NAFTA investors. In fact, ex-NAFTA FDI inflows to Mexico have actually grown faster (average annual growth rate of 23.5%) than NAFTA FDI (10.1%) during the period of 1999-2016, and accounted for 52% of total FDI in 2016. Mexico’s domestic regulation has been upgraded substantially. The transformation from a monolithic-party state-capitalist economy to a multi-party democracy and market system has reduced expropriation risk to low levels. Moreover, the government has made great strides in protecting foreign investments, gradually eliminating performance requirements (such as minimum exports and local-input requirements) on a unilateral basis since the 1980s, slashing regulation to allow the entry of foreign investments from all nationalities in the 1990s, establishing a commercial code for secured transactions in the early 2000s, and currently, implementing structural reforms to enhance competition and open up historical monopolies (telecom and energy) to foreign investors. Additionally, Mexico has an extensive network of free-trade agreements (most of which have investment chapters built on the NAFTA template) and additional bilateral investment treaties with 33 countries. 

Of course, we cannot rule out a worse scenario for Mexico if the U.S. sidesteps the WTO’s legal framework and imposes retaliatory measures to reduce its trade deficit. After all, if the Trump administration actually wants to reduce its deficit with Mexico, it is very likely that switching to trade through the WTO will not help much. However, we note that breaking with the WTO would mean taking protectionism to a more dangerous level, threatening trade relations with all other important commercial partners.

Domestic political uncertainty on the rise

In addition to the growing risk of NAFTA renegotiation overlapping with the Mexican presidential elections, increasing the probability of a deadlock situation, Margarita Zavala (a key member of the right-wing PAN party) surprised the political establishment by announcing her decision to run as an independent. In October, Margarita Zavala resigned from the PAN after 35 years of affiliation and announced that she will run for president as an independent candidate. Thus, the votes of the PAN are expected to split. 

Because there is no run-off in Mexico, a more divided center-right benefits the left candidate Andrés Manuel López Obrador (AMLO). In fact, according to the most recent polls, AMLO has widened his lead over the runner-up after Zavala’s decision to leave the PAN. 

However, the ruling party PRI also benefits from the PAN split. Recent polls show the possible PRI candidate, Miguel Osorio (Minister of the Interior), running in second place. Precisely because there is no run-off in Mexico, “tactical voting” is important: that is, as the elections approach, anti-AMLO voters will likely support whoever is best placed to beat him. Also, the PRI will count on the political machinery to defeat AMLO – it governs 16 out of 32 states, whereas AMLO’s Morena party has none.

The “Frente Ciudadano por México” (PAN/PRD alliance) is wounded, but a comeback cannot be ruled out. Ricardo Anaya (President of the PAN) will likely be the alliance’s candidate, although this is not official yet. The PAN has gained four times the voter preferences of the PRD over the past months, so it is likely that they will have more influence on the definition of the candidate, and – without Zavala – there is no one who challenges Anaya now. As in the PRI’s camp, the PAN/PRD’s candidate will be determined by a small committee of party leaders instead of undergoing primary elections. A silver lining for Anaya is that the departure of the “Calderonistas” (the furthest right and most conservative wing of the party) allows him to tilt more to the center of the political spectrum and to gain some votes there. 

Still, AMLO remains very well-positioned for the elections. Besides the fact that he has consistently led polls, the low popularity of the government suggests an anti-establishment sentiment that could help the left. Tellingly, according to the “Latinobarometro” survey, Mexico is the country in Latin America where support for democracy fell the most between 2016 and 2017, possibly reflecting the population’s frustration with the political establishment’s corruption scandals. AMLO’s campaign, in fact, is focused on exploiting the anti-corruption message. Moreover, the recent strong performance of Delfina Gómez under the banner of Morena in the elections of the State of Mexico, the country’s most-populated state (and the PRI’s historical stronghold), is a sign of AMLO’s strength.

Central Bank reacts to weaker MXN, but bias remains neutral

Amid growing external and domestic risks, the Mexican peso weakened and triggered intervention. Mexico’s FX commission, formed by the Central Bank and the Ministry of Finance, stepped up its FX swap interventions. On October 25, the FX commission announced a new issuance of USD 4 billion in FX swaps to support the MXN, increasing the utilized capacity of the USD 20 billion FX swap facility announced in February 2017 to USD 5 billion (from 1 USD billion). Specifically, the central bank, which operates the FX swap facility, issued USD 1 billion of FX swaps on October 26, and will carry out six USD 500-million auctions spaced out in one-week periods ending on December 6 to complete the USD 4 billion. Even though 25% of the FX swap facility’s capacity is now committed, the FX commission still has ample ammunition to intervene, considering the USD 173 billion (15% of GDP) international reserves and the USD 86 billion flexible credit line granted by the IMF. The key advantage of the FX swap instrument is that it allows the Mexican government to intervene in the FX market without compromising international reserves, so it can be used as a first line of defense in a scenario of higher volatility. 

Against this backdrop, two central-bank board members see a possibility of rate hikes, while the three-member majority seems comfortable with the current level of the reference rate (but does not see room for rate cuts). The monetary-policy minutes of September’s meeting (published on October 12) revealed that two board members (probably Javier Guzmán and Manuel Ramos Francia) do not rule out further rate hikes. More recently, however, during an interview for the Wall Street Journal (published on October 25), Governor Agustín Carstens talked down the recent increase of MXN volatility, by stating that “as things are going today, we are on track” [to bring down inflation]. Carstens’ last meeting as Governor of Banxico will be on November 9, before he departs to head the Bank for International Settlements (BIS). Amid Fed interest-rate hikes and growing uncertainty over domestic and foreign politics, rate cuts seem to be off the table. 

Our base case is that the policy rate will be kept at 7% at least until the beginning of 2H18. Currently, ex-ante real interest rates are somewhere between the borders of what the Central Bank considers “neutral” and “tight” (considering the neutral real rate range estimated in a recent inflation report: between 1.7% and 3.3%). As uncertainty diminishes, we think the central bank will gradually move to the center of this range, starting with two 25-bp rate cuts in the second half of 2018. 

Inflation is trending down

After peaking in August (at 6.66%), annual inflation increased again between the second half of September and the first half of October, but the diffusion index continues to indicate that inflation is becoming less generalized across the CPI basket. Headline inflation increased to 6.30% year-over-year in the first half of October (from 6.17% in the second half of September), while core inflation increased slightly to 4.75% (from 4.70%) during the same period. The decrease of core goods (tradables) inflation (6.00%, from 6.08%) was offset by an increase in core services inflation (3.68%, from 3.53% previously). However, a cleaner indicator of prices driven by domestic demand – that is, core services excluding telecom and FX-sensitive services (which includes airfares and tourism-related services) – actually decreased, to 3.92% (from 3.94%). Turning to non-core inflation (11.18%, 10.83% previously), the increase was explained by higher inflation for energy (15.86%, from 15.57%) and regulated items (8.10%, from 4.65%), which was partly offset by lower non-core food inflation (8.26%, from 9.77%). Importantly, we note that the diffusion index, which tracks the percentage of items in the CPI basket with inflation higher or equal to four, the upper limit around the target, decreased to 73.9% (from 74.3% in the second half of September), marking the fourth consecutive bi-weekly decrease.

We now expect annual CPI inflation at 5.9% by the end of 2017 (from 5.7% in our previous scenario) because inflation for core goods (tradables) is falling slower than expected. Nevertheless, we continue expecting the downward trend of inflation to accentuate in the next months. In spite of the recent MXN sell-off (over higher U.S. Treasury yields and NAFTA risks), the exchange rate appreciated 8% in the first 10 months of the year (after depreciating 19% in 2016). This will be the leading driver of disinflation, with a benign impact on core tradable prices. Furthermore, agricultural inflation, which is volatile and still running at an abnormally high level, will likely continue to show a reversion in the coming months, as it did in September and October.

Growth weakened in 3Q17

Data for 3Q17 points to an economic slowdown, which was exacerbated by the earthquakes that took place in September. The flash estimate of GDP growth features a contraction of 0.2% from the previous quarter, after posting robust sequential expansions of 0.6% in 2Q17 and 0.7% in 1Q17. The monthly GDP proxy (IGAE) grew 2% qoq/saar in August, with a deterioration in industrial production (-2.4% qoq/saar, -1.9% in July) offset by robust services sectors (4% qoq/saar, 3.6% in June). Industrial production was dragged down by the decline in mining output (-7.7% qoq/saar, -3.4% in July), largely oil, and construction activity (-3.9% qoq/saar, -3.1% in July). In fact, construction reflects the poor performance of investment, which is currently the weakest component of domestic demand. In contrast, manufacturing output gained some traction (1.1% qoq/saar, -0.1% in July). On the services side, retail sales performed poorly in August (-3.2% qoq/saar, same as in July), but the more comprehensive monthly proxy for private consumption (4% qoq/saar in July, latest print) – better correlated with GDP – continued to show positive dynamics. We believe that September’s data will likely feature poor readings for both the industrial and service sectors, because of the earthquakes. According to a recent survey by the statistics institute (INEGI), 40% of the surveyed businesses in the six states affected by the earthquakes, which together account for one-third of the economy, closed their doors for at least one day.

Given the poor flash GDP print in 3Q17, we have revised our GDP growth forecast for 2017 (to 2.1%, from 2.3%) – which would imply a slight deceleration with respect to 2016 (2.3%) – but continue expecting growth of 2.1% for 2018. We note that uncertainty over growth next year is high, considering the debate related to NAFTA and the presidential elections. However, we also believe that there will be significant buffers for activity in coming quarters. Manufacturing output will likely be boosted by the strength of the U.S. industry, as the U.S. ISM manufacturing index reached 60.8 in September (the highest level in thirteen years). Furthermore, we believe that falling inflation coupled with robust employment (growing consistently above 4% year-over-year in the first nine months of 2017) will sustain consumption growth. In fact, in spite of all the noise surrounding NAFTA, private investment is far from collapsing – it was actually up by 1.8% in 2Q17, once adjusted for calendar effects, while imports of capital goods grew 4.7% qoq/saar in 3Q17.


João Pedro Bumachar
Alexander Muller

Please see the attached file for all graphs. 

< Back