Itaú BBA - Finding MXN equilibrium in more challenging conditions

Macro Vision

< Back

Finding MXN equilibrium in more challenging conditions

February 20, 2017

In worst-case scenario, U.S. border tax would push the Mexican peso’s fair value to 23.1 to the dollar

Traditional exchange rate models indicate that the Mexican peso is undervalued. When presenting the results, we are often asked what the fair value is if the deterioration of the energy balance seen over the past few years is permanent, and how protectionism in the U.S. would alter the equilibrium exchange rate. 

To address these questions, we estimate the non-oil exports and imports as a function of the real exchange rate as well as other macro variables (internal demand, U.S. industry, etc.) and calculate the necessary depreciation to generate a non-oil surplus that is sufficient to offset the decline in energy exports and import tariffs in the U.S.

Our results show that the Mexican peso is still undervalued, even if the U.S. imposes a 20% import tariff on Mexico’s products and the loss of energy exports is permanent. However, under a border tax adjustment, the currency likely has more room to weaken.   

Fundamentals point to undervaluation

Traditional exchange rate models indicate the Mexican peso is undervalued. If we calculate Mexico’s real exchange rate as a function of the relative productivity between Mexico and the U.S., terms of trade, sovereign risk (more specifically, the 5-year CDS) and the interest rate differential with the U.S., we estimate the Mexican peso should be trading at around 18.1 to the dollar currently.

When presenting the results, we often hear two questions: what is the fair value if the deterioration of the energy balance seen over the past few years is permanent, and what is the fair value if protectionism materializes.  

However, at least some of the impact of potential protectionism in the U.S. and deterioration in the energy balance are already captured in the model. The deterioration in the energy balance is partly associated with lower oil prices (terms of trade is one of the explanatory variables). Between 2013 and 2016, Mexico’s energy balance deteriorated by around 2% of GDP. If energy export prices (mostly crude oil) and energy import prices (mostly refined products, which fell by less than crude prices) were unchanged, the deterioration would have been more modest (around 0.7% of GDP). Furthermore, the risk of protectionism and the lower net energy exports have a negative impact on sovereign risk.

In order to address the impact of lower net energy exports and protectionism on the Mexican peso, we first try to remove any effect on the currency related to these two factors from our model. Then we will treat the two factors using a separate set of equations and assumptions.  

Mexico’s 5-year CDS has been decoupled from its peers since early 2016. If there wasn’t the underperformance, the CDS would be trading currently at around 100 bps (instead of 145 bps). Now, assuming terms of trade at its long-run average (more precisely, the average of the 20-year period ended in 2015, which is 19% above the current level) and “correcting” the 45-bp underperformance of Mexico’s CDS, the fair value of the Mexican peso would be 16.5 to the dollar. This is our counterfactual scenario: one that mimics the absence of the energy balance deterioration and the threat of protectionism. 

Trade openness helps dealing with shocks

We estimate four equations (non-oil exports to the U.S., non-oil exports to other destinations, non-oil imports from the U.S., non-oil imports from other destinations). Non-oil exports to the U.S. are explained by the USDMXN real exchange rate, U.S. manufacturing production and the USDCNY real rate (as China competes with Mexico in the U.S. market, a weaker CNY hurts Mexico’s exports). Non-oil exports to other destinations are explained by LatAm GDP and by the Mexican peso against a basket of LatAm currencies (in real terms). Imports from the U.S. are a function of the USDMXN real rate, internal demand in Mexico and non-oil exports to the U.S. (consistent with the fact that a large portion of Mexico’s products sold in the U.S. use intermediate goods produced in the U.S.). Finally, non-oil imports from other destinations are explained by internal demand and by the Mexican peso against a basket of LatAm currencies.

An interesting result from this model is that the trade balance of Mexico is highly sensitive to the exchange rate, given the degree of openness of the economy. For example, a 10% multilateral real depreciation of the exchange rate lifts Mexican exports by 2.4%. At the same time, it reduces imports by 2.7% (also considering the positive impact of higher exports to the U.S. on Mexico’s intermediate goods imports). However, given the weight of non-oil exports and non-oil imports on Mexico’s GDP (33% each, approximately), these variations improve the trade balance by 1.8% of GDP.

What MXN depreciation offsets the loss of oil exports?

The next step is to calculate the necessary depreciation to generate a non-oil trade balance that offsets the worsening of the energy balance. To be more precise, the energy balance in 2016 was 2.3% of GDP lower than its previous 20-year average.

Based on these equations, we estimate the real exchange rate would need to weaken by 12% against the dollar (assuming all other currencies in the models are stable versus the USD) to generate a 2.3% of GDP improvement in the non-oil trade balance. So, without protectionism and assuming the energy balance deterioration is permanent, the Mexican peso’s fair value would be 18.5 to the dollar (starting from the 16.5 level, which is the equilibrium exchange in the counterfactual scenario described above).

The MXN facing protectionism

And what would the impact of protectionism be? Suppose the U.S. imposes a 20% tariff on Mexico’s exports. The impact on non-oil exports to the U.S. (we assume Mexico can redirect its energy exports without a significant cost) would be similar to a 16.7% appreciation of the Mexican peso against the dollar (USDMXN goes down by 16.7%, to be more precise): they would fall by 3.8%, meaning a loss of 1.0% of GDP. However, considering the U.S. content in Mexican exports, we estimate that non-oil imports from the U.S. would fall by 0.3% of GDP. So, the net impact would be 0.7% of GDP, without assuming trade retaliation measures by the Mexican government (which could make Mexico’s imports fall more). We estimate that a 4% depreciation (assuming other currencies are unchanged against the dollar) would offset the 0.7% of GDP loss. Thus, if the energy balance deterioration is permanent and a 20% import tariff is imposed on Mexico’s products in the U.S., the fair value of the peso would be 19.2 pesos to the dollar.

How about the border tax? The whole world would be affected by a border tax, so the assumption that the other currencies would stay unchanged against the dollar doesn’t hold. Let’s accept the textbook model that says that there is an appreciation of the dollar of enough magnitude to leave all the trade accounts in the world unchanged if a border tax is imposed. For example, a border tax with a corporate tax rate of 20% means that if the Mexican peso (and each of the other currencies) depreciates by 25% against the dollar (USDMXN up by 25%), all the trade accounts would be the same. In all, if the textbook model holds, a scenario of permanent deterioration of the energy balance and border tax would leave the fair value of the Mexican peso at 23.1 to the dollar. However, due to frictions and potential retaliations, the adjustment of the dollar may not be so perfect, which would have a significant impact on exports and imports to and from the U.S. The potentially disruptive effects of the border tax mean that the Mexican peso (and possibly many currencies) will likely overshoot before the exchange-rate market puts all the trade accounts back in the pre-border tax equilibrium.

Too much bad news seem to be priced in, but more weakening is possible

In all, our analysis suggests that the Mexican peso remains undervalued, even given harsh assumptions (20% tariff on Mexico’s exports to the U.S. and no reversal of oil exports with the energy reform). However, with a border tax, such as the one currently discussed in the U.S. Congress, the currency has more room to weaken. On the other hand, our exercises do not consider the local political dynamics. As the odds of a victory of a non-establishment candidate in the 2018 presidential elections rise, new hurdles for the peso could emerge. For example, the energy reform could be slowed down, as it is up to the executive power to conduct the oil auctions at a pace it determines. Also, public debt (at 50% of GDP) is already high, and rising, so potential expansionary policies may trigger a more meaningful deterioration of sovereign risk.


 

João Pedro Bumachar

Alexander Müller

 



< Back