SALINAS, TRADE AND MEXICO'S CURRENCY DEVALUATION George Baker While rumors of an overvalued peso had been circulating in Mexico and abroad for over a year, the devaluation of the Mexican peso three weeks after the installation of a new president on December 1 is surprising. With a closer look at trade statistics, it is also understandable. The first surprising element is found in the retrospective view: Looking back, despite the lack of an investment-grade rating by Moody's and Standard and Poor's, American investors put some $35 billion dollars in equities and bonds during the six-year Salinas period from 1989-94. A second surprise is that the devaluation broke one of the fundamental terms of the working agreement between the government and Mexican private industry that had been in force since December 1987. The agreement was that the government would maintain a predictable exchange rate policy as well as low prices for labor and fair prices for state-produced products and services (notably, energy), and, in return, industry would increase its investments in plants, equipment and real jobs Mexico. A third, if less visible, surprise is that former President Carlos Salinas broke one of the unwritten conventions of devaluation politics in Mexico: unlike other Mexican presidents who faced adverse economic conditions that made devaluation the lesser of alternative evils and who decided to take on investor anger and disappointment during their terms of office, President Salinas decided to pass the responsibility for devaluation to his successor. Other Mexican presidents, notably Luis Echeverr=EDa and Jos=E9 L=F3pez Portillo, decided that it was better that th= eir successors start off their terms in office with a stable currency, even if it meant a peso that was lower in value. While it is certain that Salinas's successor knew, at least by late August, that a devaluation of the peso was inevitable, it is not clear whether or not his originally intended successor, the assassinated Luis Donaldo Colosio, also knew that a devaluation would soon be necessary. In neither case is there a solid basis for speculating on the reaction of either presidential candidate to the news that the government's exchange rate and inflation and trade statistics papered over profound weaknesses in the Mexican economy. It is likely that the timing of the devaluation was the subject of intense negotiations between the Government and the sequential PRI candidates. While a complete autopsy on the inflation statistics of the Salinas period is not yet possible, owing to the lack of data on pent-up demand in distinct markets, a sketch of a road map can be drawn. The year Salinas came into office inflation was some 150%, the year he left office inflation was under 10%. While these two statistics doubtless point to a general trend, the figures do not take into account the existence of the wage-price-parity agreement, known in Mexico as the "Pacto," that had been renewed annually since December 1988. The government's measurement of inflation is primarily keyed to a basket of prices for goods whose prices are government-controlled, such as eggs, tortillas, beans, corn, public housing, gasoline and electricity. What price behavior would have been seen in the absence of government controls is for the moment anyone's guess. The emphasis on government prices had a corollary during the Salinas period, which was that in non-controlled sectors of the economy there was a tendency toward price parity with the United States. The economy came to have two price determinants: one was low income prices controlled by the government, the other was high-society prices equivalent to those of Fifth Avenue or Rodeo Drive. For the business traveler to Mexico City, Monterrey or Guadalajara, high-society prices were everywhere, from hotel rooms to restaurant tabs and attorney fees. A U.S. Embassy official estimates that to open an office in Mexico City an international company should have a half-million dollar budget. Unfortunately, such high-society prices (made higher for the foreign visitor by a progressively overvalued peso) spilled over into the tourism sector, which, counter-intuitively, made Mexico vacations out of reach for tens, or hundreds, of thousands of American and European tourists. The cost of foregone tourism exports during the Salinas years has not yet been estimated. Regarding investments, Salinas in his outgoing State of the Union Address on November 1, 1994, bragged that during his six-year rule there had been some $50 billion in private investments in Mexico. Unfortunately, such amounts were primarily invested in the stock market, and very little in new plants, equipment and jobs. Official Mexican trade statistics for the period 1989-93 tell an ambiguous story about the controversial role of the oil industry in Mexico. While the president of Mexico does not set world oil prices, he does set forth the framework in which investment in the oil sector takes place. At the insistence of Salinas, there were no private investments in any area of the petroleum sector, despite the badly needed infrastructure upgrading in production, refining and natural gas transmission and distribution. Pemex, meanwhile, was given a shoe-string investment budget and was forced to pay exorbitant taxes. Oil production stayed flat, and revenues from oil exports fell 11% during the period. To judge from official statistics, oil exports, as a percent of total merchandise exports, fell some 36%, to 14% in 1993, from 22% in 1989. These figures will be revised momentarily. Statistical appearances of the Salinas period are the most deceptive in the area of manufacturing exports, which, seemingly, increased 67%, to $42 billion in 1993, from $25 billion in 1989 (Table 1)*. To understand the needed correction, a clear idea of a "merchandise export" must be held in mind. What counts as an manufactured export is a product for which the manufacturers are responsible for its design, engineering, market strategy and whole pricing. If one of these elements is missing, the plant is performing an international assembly service, not engaging in the manufacturing of exports. Before Salinas, this distinction used to operate in Mexican trade statistics. Mexico's "in-bond industry," which in official statistics, accounted for some $22 billion of merchandise exports in 1993, is made up of plants that operate in a legal and tax framework like that of a Free Trade Zone (FTZ). The exports of these FTZ plants used to be treated as export services, comparable to those of the tourism industry. It was never considered that a product with $95 of imported components and $5 of Mexican labor, supervision and packaging should be treated as a $100 Mexican export. In early 1991 this distinction was dropped in favor of treating the FTZ export as if it had $100 of Mexican content. A second category of questionable exports is intra-firm trading. When a Ford or a Volkswagen plant in Mexico manufacturers and exports an engine or builds and exports a car to their respective parent companies, no Mexican engineering or marketing executive is centrally involved in the design, planning or market-evaluation processes. Foreign-owned, non-FTZ plants in Mexico (such as IBM's) do pay taxes on profits from operations, but 'profits' are cooked by intra-firm accounting practices, and are not the result of market successes and failures. Restated in this way, and estimating intra-firm trading as 25% of general exports (exclusive of FTZ exports, which are disallowed in their entirety), Mexican exports in 1993 were $15 billion, not $41 billion (Table 2)*. Oil exports were 40% of total merchandise exports in 1989, not 22%, a ratio that indeed fell in the succeeding years--to 30% in 1993, but not to 14% as official trade statistics suggest. By this reading, the Salinas government understated its dependence on oil exports by nearly 50% and overstated the size of Mexico's true merchandise exports by a factor of 1.8. (Tables 3 and 4). In such a make-believe world it was easy to see lose sight of the underlying reality that Mexico's growing trade deficit was pointing to, namely, that the other economic indicators were not accurately measuring the weakness of an economy based on high-liquidity investments and a trade account based non-market prices for a large share of imports and exports. One immediate outcome of the unexpected devaluation of the peso is that it all but eliminates Carlos Salinas as a viable candidate for the directorship of the World Trade Organization. This successor to GATT, the success of which will depend on the fairness of its policies and the transparency and honesty of its statistical reporting, will not be best served by having, as its first chief executive, an authoritarian political leader for whom trade statistics are instruments of politics. While the Salinas approach to trade statistics may be good enough for the executive branch of the U.S. Government and certain editors of the U.S. press, which support the Salinas candidacy, it should not be accepted by the international community, including Mexico. A second immediate outcome is that it increases the urgency with which economic planners in Mexico need to rethink the assumptions shibboleths of Salinonomics. George Baker directs Mexico Energy Intelligence, a subscription service based in Oakland, California. Note: A pre-devaluation version of the analysis of Salinas trade statistics was published in translation in Reforma (Mexico City) on Dec. 19, 1994. g.baker@energia.com Tel: 713-627-9390 Fax: 713-627-9391 ------ George Baker