September 1999
Federal Reserve Bank of Dallas

Can Mexico Weather Its Next Election Cycle?

Over the past 25 years, Mexico has suffered financial crises toward the end of its six-year presidential terms. Because problems during election years have become so common in Mexico, market participants are naturally cautious about the potential for economic turbulence over next year’s Mexican presidential election—despite efforts to ensure that things will be different this time. The big question for Mexico and, of course, Texas businesses that depend on trade with Mexico is, Can Mexico weather its next election cycle?

In this presentation I first review Mexico’s economic performance since the 1994-95 peso crisis. Second, I examine vulnerability indicators for Mexico and compare Mexico’s current economic situation, 12 months before its next presidential election, to the same period before previous election periods. Next, I assess the likelihood of a crisis over the coming year given political and economic changes that have occurred during the last five years. Finally, I provide an economic outlook for Mexico through the year 2000. The bottom line of the presentation is that developments in Mexico since 1995, especially the adoption of a flexible exchange rate regime and better debt management, make economic turbulence in 2000 less likely than in the past.

Mexico’s Economic Performance
Overall, the Mexican economy has done remarkably well over the last four years. As shown by the bars in Figure 1, real GDP growth averaged over 5 percent from 1996 to 1998, and real GDP per capita, shown by the line, now exceeds its pre-crisis level by about 4 percent.

Although last year’s Russian crisis stalled the economy in the fourth quarter of 1998, growth resumed in the first quarter of this year and has shown strength in the second quarter, growing by nearly 8 percent at an annual rate.

Inflation, which rose over 100 percent at an annual rate in early 1995, declined to less than 15 percent in 1997. Although inflation moved up to nearly 19 percent last year, largely due to a sharp depreciation of the peso, the Bank of Mexico has since been able to reverse that trend with tighter monetary policy.

Short-term domestic interest rates, the blue line in Figure 3, which had been falling since 1995, peaked at more than 40 percent in 1998 after the Russian crisis. Over the same period, the peso depreciated 26 percent against the dollar. Domestic interest rates started falling and the peso strengthened in the fourth quarter of 1998, although those trends were temporarily reversed when the Brazilian crisis struck in January this year.

But despite what many had originally thought would be a lingering period of volatility, the Brazilian crisis resulted in only a relatively minor setback on Mexico’s path to recovery. The fallout from Brazil’s devaluation was limited by two factors: first, by Mexico’s willingness to raise interest rates quickly in the face of a sharp fall in the peso and consequent inflationary pressure, and second, by precautionary adjustments that investors already had made during the Russian turbulence six months earlier.

Vulnerability Indicators
As Brazil has stabilized, Mexico has benefited from lower interest rates and a renewed inflow of capital to the region. While Mexico is expected to continue to grow in the remainder of 1999, the question remains, Can Mexico weather its next election cycle? The past 25 years of Mexican history is littered with economic crises that tend to fall during election years.

As Figure 4 indicates, major devaluations and financial crises have followed most Mexican elections since 1976. This figure shows the path of the nominal exchange rate from 1973 to the present, with election years indicated by the shaded bars. In three of the last four elections—1976, 1982, and 1994—a major devaluation and economic crisis occurred around an election. The 1988 election was an exception to the rule as Mexico had not fully recovered from its 1982 crisis and the peso depreciated earlier under high inflation and mid-term turbulence following oil price declines and its 1985 earthquake.

The question is: Why do crises occur in election years? Election years have led to a confluence of forces that make the economy vulnerable to crisis. Incumbent governments in Mexico, like those in many other countries, have the incentive to keep the economy growing in an election year in order to attract as many votes as possible. Going into an election year, the government tries to sustain or increase fiscal spending. Monetary policy is kept loose and the pace of lending to the public and private sectors is maintained. This causes inflationary pressures to build up and, given a highly managed nominal exchange rate, leads to appreciation and overvaluation of the real exchange rate. An overvalued real exchange rate reflects a drop in international competitiveness and leads to reduced exports and increased imports. This, in turn, generates a widening current account deficit and increased speculation of an impending devaluation, which lowers international reserves as capital flees the country. Eventually, under mounting pressures, a balance-of-payments crisis erupts and the currency is devalued.

In order to assess whether Mexico will fall into another election year crisis, let’s examine major vulnerability indicators and compare their current behavior to past election cycles. Vulnerability indicators attempt to measure the susceptibility of an economy to a crisis. The indicators I examine include the growth in real government expenditure (an indicator of fiscal imbalance), the growth in domestic credit relative to GDP (an indicator of monetary stimulus and inflation pressures), real exchange rate appreciation (a measure of how internationally competitive the country is), and, finally, the current account balance (an indicator of how reliant the country is on foreign capital inflows to fund imports of goods and services). In general, countries with high growth in government expenditure, rapid expansion of domestic credit relative to GDP, overvalued real exchange rates, and large current account deficits are susceptible to a financial crisis.

Figure 5 shows the growth of real government expenditure, which is fiscal spending deflated by the price level. The dashed horizontal line indicates the average value of real government expenditure over the sample period, while the shaded bars indicate election years. The squares on the plotted line are placed at the year before a balance of payments crisis. The average value of the growth of real government expenditure the year before crises is shown with a shaded square on the right axis. 1988 is not classified as an election year crisis because a major devaluation did not occur in that year.

As you can see, in the years prior to elections, fiscal spending accelerates. On average, as indicated by the shaded square on the right, the growth in real fiscal spending was 18 percent before crises. In contrast, real fiscal spending is projected to increase by only about 3 percent from 1998 levels.

The increase in fiscal spending prior to crises was usually accompanied by an acceleration in domestic credit relative to GDP, a measure of monetary stimulus and price pressures. Monetary expansion was especially evident in 1982, and to a lesser degree in 1976 and 1994. On average, as shown on the right, the year before crises, domestic credit relative to GDP grew by about 5 percent, while so far this year it has fallen 5 percent.

The accumulated price pressures prior to crises, stemming from fiscal and monetary expansion, along with a highly managed exchange rate, generally led to an appreciating real, inflation-adjusted, exchange rate, as indicated in Figure 7. The real exchange rate appreciates when domestic prices increase faster than foreign prices and cause the country to become less competitive internationally. On average, the real exchange rate was about 26 percent higher than its long-run average the year prior to crises. Notice that in 1988, the only election year that didn’t suffer a crisis, the real exchange rate was below its long-run average. Currently, it is only about 7 percent above its long-run average.

Reflecting all of these vulnerabilities, international trade and funding imbalances generally deteriorate in the run-up to balance-of-payments crises. As shown in Figure 8, the average current account deficit prior to crises was 4.4 percent of GDP. It is now about 2.5 percent, slightly below its long-run average, but much better than past periods leading up to crisis.

Crisis Assessment
Overall, Mexico’s current situation appears to be better than past years leading up to crises. But, there is quite a wide variation across episodes. Moreover, history shows that there is still a possibility that these indicators can worsen over the next 12 months.

Several key factors, however, have changed since the 1995 crisis that would suggest that turbulence is less likely now than in the past. These include a floating exchange rate, which can limit real exchange rate overvaluation; a weak banking sector, which is unlikely to be a source of rapid credit expansion; political competition, which may restrain fiscal spending; and better debt management.

As shown in Figure 9, since Mexico floated its exchange rate in December 1994, the nominal exchange rate has become more volatile—indicated by the bars on the left—as one would expect under a floating regime. At the same time, the volatility of the real exchange rate—shown by bars on the right—has declined. Increased flexibility in the nominal exchange rate has acted as a shock absorber to external turbulence, permitting adjustments to changing pressures rather than allowing them to build up. Consequently, the real exchange rate has become less overvalued and not as volatile, which has decreased the likelihood of a large discrete devaluation. Much the same happened, for example, during Mexico’s period of exchange rate flexibility in the years prior to its 1988 election, and was probably an important factor in averting a major devaluation that year.

Bank credit expansion was a source of vulnerability in past crises, particularly in 1994. Historically, the banking sector has amplified the boom and bust cycle by extending easy credit to marginal borrowers when asset prices were high and then withdrawing that credit when asset prices fell. In 1994, before the peso crisis struck, rapid bank lending fueled unsustainable spending.

Recently, however, the banking sector is less a factor in excessive growth simply because banks have never recovered from the peso crisis. Large corporations still have access to international capital markets, but individuals and small businesses are credit constrained. The level of real bank loans outstanding has continued to fall since the peso crisis, while nonperforming loans as a share of total loans, not shown, is still relatively high at about 15 percent.

Over the last 14 years the political dynamics in Mexico have dramatically changed, with mixed implications for the coming year. As Figure 11 shows, Mexico has moved from a one-party system where the government was completely dominated by the PRI to a multiparty system where power is less concentrated. As a result, elections in the year 2000 are likely to give the PAN, the conservative and second most popular party, the best chance it's ever had to win the presidency. The PRD, the most liberal of the top three parties, has little chance of winning the election by itself and is unlikely to join a coalition with the PAN. Greater political contestability has induced the PRI to become more aggressive in producing a candidate who can win elections. In a major change from its tradition of the president hand-picking his successor, the PRI is allowing a primary where the candidate will be chosen by a popular election. This is likely to produce a candidate who is more of a populist than a technocrat. The economic implications are that increased political rivalry between parties may generate pressures for more election year spending. However, because power is now shared in the Congress, there are more checks in the system that will likely limit excessive election year spending.

Another factor tending to bolster Mexico’s stability is the shift away from reliance on volatile portfolio capital inflows, which is investment in the stock and money markets, to more stable foreign direct investment, which is property, plants and equipment. Figure 12 shows that portfolio investment, the blue line, is much more volatile than direct investment, the red line. Foreign direct investment has grown relative to portfolio investment since 1997, and, consequently, its importance for funding the current account deficit has grown as well. In the most recent four quarters, foreign direct investment corresponded to 75 percent of the current account deficit. Given our projections of a current account deficit of $12 billion in 1999, foreign direct investment should account for about 80 percent of current account financing needs.

Figure 13 shows external debt service as a share of international reserves, which is a gauge of the ability to service existing debt should international capital markets dry up. In other words, it’s a country’s foreign debt payments against current liquid foreign assets which is a measure to assess if debt payments can be made if that country no longer has access to foreign capital.

As shown here, Mexico’s debt service as a share of reserves has fallen markedly since 1994 to a level which, by international standards, is relatively low.

The external debt service projection reflects the new financial package Mexico announced on June 15, 1999. Mexico will receive a $4.1 billion standby agreement from the IMF, which essentially rolls over the remainder of Mexico’s 1995 IMF loan until 2001. In addition, loans of $5.2 billion from the World Bank and $3.5 billion from the Inter-American Development Bank will allow Mexico to ease pressures on public sector funding. These loans, in addition to a $4 billion credit line from the U.S. Export-Import Bank, will cover nearly 80 percent of the public sector’s expected external debt service in 2000, and provide about 35 percent of the country’s total external debt service. The $6.8 billion swap line associated with the North American Framework Agreement is a potential source of assistance should Mexico fall into crisis.

Mexico’s Economic Outlook
Figure 15 summarizes our outlook for Mexico. Economic activity is projected to continue its moderate recovery, based on declines in inflation and interest rates, stable oil prices, and continued access to international capital markets. The real appreciation of the peso seen so far this year is expected to slow. CPI inflation is projected to decline from 18.6 percent in 1998 to just under 12 percent in 2000. The narrowing of the current account deficit this year due to weak consumer spending is anticipated to be reversed next year as domestic demand is expected to rise in the coming election year.

There are, however, some significant downside risks to the forecast, especially given the uncertainties surrounding election year politics in Mexico. Risk factors include a larger than expected deterioration in the fiscal balance, due to either increased fiscal spending or weaker oil prices; a slowdown in the U.S. economy, which could be accompanied by a U.S. stock market correction; or significant increases in U.S. interest rates. Also, a major reversal of progress toward stability in Argentina or Brazil that destabilizes international financial markets could set back Mexico’s recovery. Overall, however, we are cautiously optimistic that Mexico is better positioned this year, compared to previous election cycles, to weather most storms on the horizon.

—David M. Gould

About In Depth This article is based on a presentation by David M. Gould, senior economist and policy advisor, Research Department, Federal Reserve Bank of Dallas. The views expressed are those of the authors and do not necessarily reflect the positions of the Federal Reserve Bank of Dallas or the Federal Reserve System.