The Mexican peso has depreciated greatly in the past few weeks with respect to the dollar, which would imply a strong undervaluation of the peso (unless the government deficit and a lax monetary policy could explain a good part of its weakened state), far from target conditions for the economy.

Without a doubt, international volatility, the monetary policies of the principal economies of the world, price fluctuations of raw materials, and the electoral process in the United States have all played a role in a fluctuating exchange rate. The Mexican peso, along with many other currencies, has been under strain since 2011 when the then-governor of the U.S. Federal Reserve announced the beginning of the end of monetary easing and when his successor, Janet Yellen, signaled the possible beginning of interest rate increases. Up until August of 2015, the anticipation of changes to U.S. interest rates was the principal variable in explaining fluctuations. Since then, prices of raw materials, especially oil, have played a more relevant role, spurred on by difficulties in China. In the last few weeks, the probability of Donald Trump winning has also begun to increasingly weigh in.

Even so, internal economic policy has also contributed. It is highly probable that the peso would have been stronger today if it had greater credibility in terms of public finance, and if adjustments for a downward trend in debt to gross domestic product were already underway.

The peso can depreciate even as a result of mere anticipation of future events based on the probability of whether interest rates increase or Trump wins. This results in an opportunity for arbitrage that puts pressure on the peso and pre-empts the future event in the following way: If higher interest rates are expected or Trump is elected, the peso loses value, so it makes sense then for one to incur debt in pesos and to repay the debt once it costs less to acquire them, considering the depreciation. These “short-term” operations explain a part of the process of depreciation in the recent weeks and months. The questions that need to be asked are how much of the fluctuation is a result of this and how authorities should confront the challenge posed by this uncertainty.

In order to estimate the impact, a series of regressions were conducted by making daily observations of the exchange rate of the dollar for various countries between July 27 and Sept. 19, 2016 (as dependent variables), and the probability of an increase in interest rates during the Federal Reserve meeting on Sept. 21, the daily average of polling for Trump, and the international price of oil as independent variables.

The results that were obtained are very interesting. The currencies whose fluctuations are better explained by these variables are, in descending order of explanatory power: Colombian peso, rand, yen, Singapore dollar, ringgit, Australian dollar, Mexican peso, Chilean peso, yuan, and real. However, only for the real, ringgit, Australian dollar, yen, and Mexican peso, in that order, does the Trump effect generate more explanatory power. For the euro, ruble, Colombian peso, and Canadian dollar, the most important variable is the price of oil, while for the yuan, Chilean peso, Singapore dollar, and rand, the most important factor is the probability of an increase in interest rates.

It is really interesting to consider, contrary to intuition and to the coverage that the exchange rate has received, that the Mexican peso is not the currency that most greatly reacts to these three variables, individually or collectively. However, there are only three currencies that react to all three of the variables: the yen, Australian dollar, and Mexican peso. It is precisely the fact that the peso reacts so negatively to a higher probability of increased interest rates, improvements in Trump’s polling, and falls in oil prices that make it vulnerable. Japan and Australia also share this triple influence but, unlike Mexico, are not considered developing countries, so a structural threat as a result of these fluctuations in the exchange rate is not even considered a concern for them. In this way, Mexico is special.

The government can take six measures to confront this exchange rate risk and the contamination of the rest of the economy. One, the Bank of Mexico can increase benchmark interest rates; it’s already done it on two occasions and the markets expect that it will happen again next week (in a more pronounced way) even though the Fed announced yesterday that it will not increase its rate in September, but that it will keep the option in mind for November or December. The lack of an increase in the U.S. rate is both good and bad news. Good news, in the sense that pressure on the peso, per this decision, will diminish for the time being. Bad news, given that the indecisiveness of fiscal authorities in Washington guarantees uncertainty moving forward and can complicate things toward the end of the year, above all if the election favors Trump. Increasing the interest rate in Mexico increases the cost of debt, which makes it less attractive to use pesos for arbitrage operations. This strategy, however, has limits and involves an economic cost, given that the cost of credit in all areas increases and not just the exchange rate.

Two, the government is obligated to send a message of fiscal consolidation to the markets and to not dally in doing so. The fiscal package sent to Congress on Sept. 8 is insufficient. In order to minimize the negative effect of fluctuations in oil prices on the peso, it is necessary that the government’s fiscal plan is achieved in 2016 and that a reduction in debt-to-GDP is apparent in 2017. A robust fiscal plan is the only antidote to the risk posed by the price of oil and to prevent Mexico from being treated as a macro-economically developing nation.

Three, it’s important that the possibility exists for an abrupt revaluation so that hedge funds run a risk upon conducting short-term investments in pesos. This could involve refraining from accumulating international reserves during periods of revaluation and also unexpectedly injecting dollars into the market in those moments.

Four, it is important to send a message to the market in terms of an economic sense of direction, and the best way to do so is the approval of the Trans-Pacific Partnership before the election in the United States, and to do it alongside Japan, Australia, New Zealand, Singapore, and others.

Five, Wall Street should be educated so that it knows the cost of a renegotiation of the North American Free Trade Act for the United States in terms of competitiveness and tariffs on its exports (surely higher than those Mexico would face), and so that the perceived probability of a serious commercial problem is lessened.

Finally, in order to safeguard the peso against a large amount of uncertainty in 2018, it is best to change our constitution to permit a second presidential term in that year’s election.