sign in sign up
<<
Home , Crisis of 1982, Economy of Mexico, Mexican peso, Mexican peso crisis, Peso

J ULY /AUGUST 1996

Christopher J. Neely is a research economist at the Federal Reserve Bank of St. Louis. Kent A. Koch provided research assistance.

This article examines the relationship The Giant between NAFTA and the of December 1994. First, the provisions of Sucking Sound: NAFTA are reviewed, and then the links between NAFTA and the peso crisis are Did NAFTA examined. Despite a blizzard of innuendo and intimation that there was an obvious Devour the link between the passage of NAFTA and ? the peso , NAFTA’s critics have not been clear as to what the link actually was. Examination of their argu- Christopher J. Neely ments and economic theory suggests two possibilities: that NAFTA caused the Mex- t the end of 1993 was touted ican authorities to manipulate and prop as a model for developing countries. up the value of the peso for political rea- A Five years of prudent fiscal and mone- sons or that NAFTA’s implementation tary policy had dramatically lowered its caused capital flows that brought the budget deficit and inflation rate and the peso down. Each hypothesis is investi- government had privatized many enter- gated in turn. prises that were formerly state-owned. To culminate this progress, Mexico was preparing to enter into the North American NAFTA Free Trade Agreement (NAFTA) with NAFTA grew out of the U.S.–Canadian and the . But less than Free Trade Agreement of 1988.1 It was a year later, in December 1994, investors signed by Mexico, Canada, and the United sold their peso assets, the value of the Mex- States on December 17, 1992. The legisla- ican peso plunged 50 percent against the tures of those countries ratified NAFTA, U.S. , and Mexico was forced to bor- and the agreement took effect on January row from the International Monetary Fund 1, 1994. The treaty substantially lowered (IMF) and the United States to get through national barriers to trade and investment a financial crisis. In 1995, inflation in Mex- in North America, giving consumers more ico soared to 50 percent and real gross do- choices and lower prices. In addition, the mestic product (GDP) fell by 4 percent. changes began to lower the cost of produc- Politicians and commentators like Ross Perot, Pat Buchanan, William Greider, and tion and to funnel investment and labor to Robert Kuttner blamed the enactment of their most productive uses. Not surpris- NAFTA for the devaluation of the peso and ingly, the costs—real and imagined—of the ensuing economic turmoil in Mexico, this reallocation of resources stirred the with some calling for its renegotiation or passions of those opposing the agreement. even repeal. As the members of NAFTA The trade provisions of NAFTA were consider expanding to encompass other designed to reduce tariffs and nontariff Latin American nations, such as , in- barriers—such as quotas and import vestors and policymakers should under- licensing—radically over 15 years. Some tariffs were reduced immediately, whereas stand the link between NAFTA and the 1 See Hufbauer and Schott peso crisis well. Did NAFTA cause or exac- other reductions will be phased in over a (1993), Aguilar (1993), or erbate the devaluation of the peso? Or did period of 10 years—15 years for certain Tornell and Esquivel (1995) NAFTA help alleviate some of the conse- sensitive sectors, such as and for more discussion of NAFTA’s quences of the crisis? and apparel. provisions.

F EDERAL R ESERVE B ANK OF S T. LOUIS

33

J ULY /AUGUST 1996

For the United States and Mexico, the Despite the impressive achievements trade provisions of NAFTA are expected to of the negotiators in crafting such a far- have their most important effects on the reaching trade agreement, NAFTA’s direct automobile, and apparel, and agri- economic benefit to the United States will cultural sectors. In agriculture, U.S. and likely be small. One representative esti- Mexican quotas were immediately con- mate of NAFTA’s annual benefits to Mex- verted into equivalent tariffs and those ico and the United States arrives at ap- tariffs will be phased out over 10 to 15 proximately the same figure for each years. As Hufbauer and Schott (1993) country;5 however, this amounts to about note, this is a remarkable achievement 0.3 percent of 1993 U.S. GDP but more given the difficulties encountered by than 5.0 percent of Mexico’s output. Schott other free trade agreements on agricul- (1994), Tornell and Esquivel (1995) and tural issues. others have argued that the most impor- Given the fierce fight in the United tant aspect of NAFTA’s passage for the States over the agreement, it is ironic that Mexican economy is that it would cement NAFTA required more substantial changes the other economic reforms in place. Krug- in Mexican law—both trade and invest- man (1993) and Orme (1993) both contend ment law—than it did in U.S. law. Average that NAFTA is most important to the United U.S. tariff levels on Mexican goods were States as a tool of foreign policy, to encour- already quite low—just four percent—on a age Mexican economic and political reform. value-weighted basis, before the introduc- tion of NAFTA.2 Mexican tariffs were higher, averaging 10 percent on imports NAFTA AND THE VALUE from the United States. Through NAFTA, OF THE PESO Mexico also committed itself to address This section lays out the case that the other long-standing U.S. concerns, like the peso was kept overvalued because of the pol- protection of intellectual property rights itics of NAFTA and then investigates whether and reform of Mexico’s regulation of for- this argument is consistent with the facts. eign investment. NAFTA was the culmination of a sig- The Case That the Peso’s Value nificant break with Mexico’s protectionist Was Artificially Inflated Because 3 past. Until the 1970s, Mexico followed a of the Politics of NAFTA 2 See Tornell and Esquivel policy of import substitution industrializa- tion that mandated highly protected mar- The most common hypothesis linking (1995). Changes in value- NAFTA to the peso crisis is that the politics kets for manufactured goods. In that weighted tariff schedules can of NAFTA caused the to decade, preliminary reforms in the direc- be misleading, however, be- systematically manipulate the value of the cause there are also some tion of freer trade were taken. The debt peso to increase support for the treaty, both quantitative restrictions. reversed that trend; for a before NAFTA was passed in the United 3 See Kehoe (1995) for a re- short period in 1982–1983, Mexico was States and during its first year. There are view of Mexico’s recent trade one of the most protected economies in two versions of this hypothesis. The first history. the world. During the de la Madrid admin- version suggests that the value of the peso istration (1982–88), Mexico took impor- 4 The GATT was an international was deliberately manipulated to secure po- organization to negotiate free tant steps to move toward more liberal litical support for NAFTA and that the de- trade among its members. It trade. Mexico lowered tariffs and joined valuation—to obtain a trade advantage— has been superseded by the the General Agreement on Tariffs and was planned well in advance. The second Trade (GATT) in 1986.4 Mexico took fur- version is less sinister. It suggests only that (WTO). ther unilateral steps toward free trade as the Mexican authorities were sensitive to 5 Krugman (1993) and Brown, part of the Salinas administration’s U.S. politics in setting exchange rate policy Deardorf and Stern (1992) dis- (1988–94) program of economic reform. after NAFTA was passed. The following sec- cuss estimates of the gains This period is known as la apertura (the tions lay out the arguments behind each from NAFTA. opening). version of this hypothesis.

F EDERAL R ESERVE B ANK OF S T. LOUIS

34

J ULY /AUGUST 1996

Deliberate Manipulation and Planned of the evidence at that time suggested that Devaluation. the treaty had cost American jobs. So there was considerable pressure to produce evi- “... the devaluation of the peso dence that showed that NAFTA would cre- had been planned for more than a ate jobs in the United States. year and was openly discussed at The Mexican government was not im- the highest levels of the Mexican mune to such pressure. In 1993, passage government. It was also widely of NAFTA by the U.S. Congress was the known in Washington. I discussed main policy concern of the Mexican ad- it in my testimony before the ministration [see Tornell and Esquivel House Committee on Small Busi- (1995)]. In August of that year, President ness in March, 1993—eight Salinas even promised to raise the Mexican months before the North Ameri- minimum wage to alleviate U.S. fears of can Free Trade Agreement was cheap Mexican labor driving down U.S. passed into law.” wages and taking jobs. Critics charge that Ross Perot, Los Angeles Times, because of such political considerations, 6 January 4, 1995. the Mexican government deliberately kept Critics like Ross Perot argue that the the peso overvalued throughout 1993 and Mexican government and the Bank of Mex- 1994 and planned the eventual devalua- ico kept the value of the peso artificially tion well in advance. high to increase political support for the Sensitivity to U.S. Politics. A more rea- treaty in the United States by creating a bi- sonable hypothesis is put forward by Ve- lateral trade surplus with Mexico. The lasco (1995) and others. They suggest United States did have a trade surplus with only that, after NAFTA was passed, the Mexico in the early 1990s. A study by Mexican authorities were sensitive to the Hufbauer and Schott (1993) was frequently U.S. political situation and may therefore cited by NAFTA proponents to support the have been more reluctant to permit the questionable notion that the growth of this trade surplus would create 170,000 jobs in peso to depreciate than they would other- the United States. The Clinton administra- wise have been. Specifically, in March tion used these arguments to sell NAFTA to 1994, the peso came under speculative the U.S. Congress primarily as a jobs pro- pressure in the wake of the assassination gram, rather than as a trade agreement that of Luis Donaldo Colosio, presidential can- would promote greater choice and lower didate of the ruling Revolutionary Institu- prices for consumers and greater efficiency tional Party (PRI). At that time, a number in production. of observers warned that the peso was overvalued and that a faster devaluation “We will make our case as hard was warranted. Velasco suggests that be- and as well as we can. And, though cause such a course of action threatened to the fight will be difficult, I deeply create political problems with the United believe we will win. And I’d like States, political exigencies may have pre- to tell you why. First of all, because vented an earlier, milder correction to the NAFTA means jobs. American jobs, value of the peso that would have avoided and good-paying American jobs. If I the drastic correction of the later crisis. didn’t believe that, I wouldn’t sup- this agreement.” Evaluating the Case that the Peso’s President at the signing Value was Artificially Inflated of NAFTA Side Agreements on 6 See also, columnist Robert September 14, 1993. Because of the Politics of NAFTA Kuttner, January 22, 1995, Critics argue that NAFTA provided the in the Akron Beacon Journal President Clinton even talked about impetus for the Mexican monetary author- and author William Greider in leaving NAFTA after three years if a review ities to maintain the value of the peso in Rolling Stone, March 9, 1995.

F EDERAL R ESERVE B ANK OF S T. LOUIS

35

J ULY /AUGUST 1996

excess of its equilibrium value. The au- barrel of oil costs $20 in the United States thorities allegedly knew that the peso was and 80 in Mexico, the law of one overvalued but gambled that this overvalu- price predicts the nominal exchange rate ation could be maintained long enough to will be $0.25 per peso. This condition secure NAFTA’s passage in the United must approximately hold, or people could States. Thus, this hypothesis requires that: make money by buying oil in the country where it is cheap and selling it in the 1. The peso was overvalued. country where it is expensive. Such arbi- trage would tend to drive the price of oil 2. The Mexican authorities knew that it down in the country where it is expensive was overvalued. and raise the price in the country where it is cheap, until the law of one price ap- 3. The Mexican authorities kept it overval- proximately holds. ued to increase or at least maintain sup- If the law of one price holds for port for NAFTA in the United States. each good in a price index and the weights in the price index are the same Although it is not possible to test the ele- for each country, then consumption bas- ments of this hypothesis, they may be ex- kets should also sell for the same price amined to see whether they are consistent when expressed in the same . with the facts. This section argues that This is called absolute purchasing power though the evidence favors the view that parity (PPP), which can be expressed as the peso was overvalued, this was not obvi- follows: ous at the time. Further, to the extent that U.S. = MEX × the peso may have been overvalued, the pIndex(t) pIndex(t) e(t), overvaluation was a result of the disinfla- U.S. tion strategy of the Mexican authorities, where pIndex(t) is a measure of the price MEX rather than a result of NAFTA. level in the United States and pIndex(t) is the analogous measure for Mexico. Of course, Nominal and Real Exchange Rates. When because of different patterns of consump- discussing the value of the peso, it is im- tion across countries, the presence of non- portant to distinguish between the nominal traded goods and differentiated goods, ab- exchange rate, or the price of a peso in solute PPP does not describe the relation of terms of , and the real exchange price levels and exchange rates very well. rate, the price of Mexican goods in terms A less stringent, but more realistic re- of U.S. goods. This section explains the re- lation is relative PPP. It says that differ- lationship between prices and exchange ences in countries’ inflation rates should rates and why the real exchange rate is the be reflected in changes in the exchange relevant measure of the proper value of the rate, so that peso. ∆ U.S. − ∆ MEX = ∆ Exchange rates and prices are linked pIndex(t) pIndex(t) e(t), through the law of one price, which says that identical goods should sell for the same where ∆ stands for the percentage change in price when expressed in terms of the same a variable over time. This equation says that currency.7 In the case of oil, for example, if inflation in Mexico exceeds inflation in the United States, the exchange rate will fall U.S. = MEX × poil (t) poil (t) e(t), to reflect the difference. That is, the peso will depreciate. Why? If Mexican goods be- U.S. where the variable poil (t) is the price of oil come more expensive than U.S. goods, con- 7 Barriers to trade, transportation in dollars in the United States at time t, sumers in both the United States and Mex- MEX costs, and imperfectly competi- poil (t) is the price of oil in pesos in Mex- ico will tend to buy more U.S. goods. This tive markets may prevent the ico at time t, and e(t) is the exchange rate will cause the peso to depreciate until Mexi- law of one price from holding. in dollars per peso. In other words, if a can goods are competitive again.

F EDERAL R ESERVE B ANK OF S T. LOUIS

36

J ULY /AUGUST 1996

A strict interpretation of relative PPP Figure 1 says that the real exchange rate, or the price of Mexican goods in terms of U.S. Index of the Real Exchange Rate (WPI) goods, should be constant over time. At Mean = 100 time t, the real exchange rate (q(t)) can be expressed as follows: 140 pMEX (t) × e(t) 120 = ᎏᎏIndex q(t) U.S. . pIndex(t) 100

For practical purposes, however, relative 80 PPP is interpreted to mean that the real exchange rate should tend to come back to 60 its historical average rather than be con- stant over time. Empirical studies suggest 40 that this interpretation of relative PPP is 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 Year useful for thinking about long-run tenden- cies in exchange rates, especially when in- flation rates are high. of the real exchange rate. If the Mexican Why is the real exchange rate impor- inflation rate minus the U.S. inflation rate tant? A currency has value only because of exceeds the rate of depreciation of the what it can purchase. The real exchange peso—that is, if rate adjusts the nominal value of a cur- ∆ MEX − ∆ U.S. > −∆ rency for its purchasing power and so de- pIndex(t) pIndex(t) e(t), termines competitiveness in world mar- kets. For example, a rise in the real —then the real exchange rate rises and exchange rate (as defined previously) Mexican goods became more expensive in means that the price of Mexican goods in terms of U.S. goods; the peso becomes terms of U.S. goods has risen. The price of overvalued. Historical measures of the cor- Mexican exports to the United States rises, rect value of the real exchange rate are im- hurting Mexican exporters, but imports perfect, though. The proper value of the from the United States become cheaper to real exchange rate can change over time Mexican consumers. Therefore the rele- because of changes in productivity, prefer- vant measure of the value of the peso is ences, legal capital controls, or other fac- the value of the real exchange rate. tors. These changes are usually slow, how- ever, leaving historical measures useful. Was the Peso Overvalued? Armed with the Respected economists like Dornbusch concept of the real exchange rate, it is still and Werner (1994) argued during 1993 difficult to determine whether the peso was and 1994 that the peso was overvalued be- correctly valued because the real exchange cause an index of the real exchange rate, rate changes over time. In the case of a as measured by the Wholesale Price Index pegged exchange rate system like Mexico’s, (WPI), was high by historical standards. a real exchange rate is functionally overval- As illustrated in Figure 1, this index rose ued or undervalued if the nominal exchange steadily from a level of 70 in 1987 to a rate is likely to be forced to change quickly. peak of about 130 at the end of 1993. By That is, the real exchange rate should be this measure, Mexican goods had become compatible with the commitment to the almost twice as expensive in terms of U.S. pegged nominal exchange rate. goods from 1987 through 1993 and the Relative PPP suggests a practical mea- real value of the peso was 30 percent sure of whether the current real exchange higher than its historical average from rate is likely to be consistent with the peg: 1975 through 1993. Dornbusch and whether it is in line with historical values Werner cautioned early in 1994 that

F EDERAL R ESERVE B ANK OF S T. LOUIS

37 J ULY /AUGUST 1996

Figure 2 Carstens (1995) argue that unit labor costs are a better way to compute real exchange Index of Unit Labor Costs rates because they more closely reflect the Multilateral Real Exchange Rate relative cost of production in Mexico and Mean = 100 abroad. Further, the real value of the peso 240 for Mexico’s trade depended not only on its value vs. the dollar, but also on its 200 value vs. Mexico’s other trading partners, and therefore they suggest that multilateral 160 measure of the real exchange rate is more appropriate. Figure 2 shows that, by the 120 beginning of 1994, the multilateral effec- 80 tive real exchange rate index, as measured by unit labor costs, had also risen substan- 40 tially—about 60 percent—since 1987 but 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 was still as low as it had ever been before Year 1986.8 In fact, it was still slightly below its historical average for the period 1975–94. Thus Gil-Diaz and Carstens argued that Figure 3 this historical measure did not show the real exchange rate to be overvalued. International Reserves Second, because the proper value of Billions of U. S. Dollars the real exchange rate can change over 35 time because of productivity changes 30 and other factors, the Mexican authorities disagreed with Dornbusch and Werner 25 about the relevance of historical mea- 20 sures. They asserted that NAFTA and 15 other economic reforms had raised pro- ductivity and had increased the correct 10 (equilibrium) value of the real exchange 5 rate; that is, the equilibrium price of 0 Mexican goods had risen. 1988 1989 1990 1991 1992 1993 1994 1995 Year “The were justifiably proud of the progress they had made in bringing down inflation, this situation was untenable and the by means of the exchange rate peso should be permitted to depreciate link to the dollar, and did not faster. want to lose it. I suspect they The hypothesis that the Mexican au- thought they were in a new world, thorities deliberately manipulated the as a result of the economic liberal- value of the peso requires that these au- ization and NAFTA.” thorities knew the peso was overvalued. Economist Jeffrey Frankel, Did they know this? In responding to Statement to the U.S. Senate Com- Dornbusch and Werner, economists at the mittee on Banking, Housing and Bank of Mexico contended that the real Urban Affairs, March 9, 1995. exchange rate was not overvalued for sev- eral reasons. First, another measure of the Also, there was very little pressure on 8 Data for the multilateral real real exchange rate—using unit labor costs the peso before March 1994, indicating exchange rate were taken from instead of price indices—did not show the that the markets did not believe that the Gil-Diaz and Carstens (1995). peso to be overvalued. Gil-Diaz and peso was overvalued. In fact, the Bank of

F EDERAL R ESERVE B ANK OF S T. LOUIS

38 J ULY /AUGUST 1996

Table 1 Mexican Consumption, Savings, Output, and Inflation

1987 1988 1989 1990 1991 1992 1993 1994

Total Consumption* 89.3 91.4 90.5 89.2 90.7 92.8 93.5 95.9 Private Consumption 78.8 81.3 80.8 79.8 80.6 81.4 81.2 82.3 Public Consumption 10.5 10.1 9.7 9.5 10.1 11.4 12.3 13.6 Total Saving* 10.7 8.6 9.5 10.8 9.3 7.2 6.5 4.1 Private Saving NA 7.8 7.5 6.6 5.1 3.8 NA NA Public Saving NA 0.8 2.0 4.2 4.2 3.5 NA NA

Real GDP Growth 0.0 1.3 3.3 Ϫ0.9 9.3 2.8 0.4 3.8 Inflation (CPI) 159.2 51.7 19.7 29.9 18.8 11.9 8.0 7.1

* Table entries are expressed as a percentage of National Disposable Income. Source: OCED National Accounts and DRIINTL.

Mexico had to intervene in the market to could to a crisis.10 These economists ar- sell pesos/buy dollars to keep the value of gued that the peso had become overvalued the peso down in January 1994, accumu- because Mexican officials had used the lating foreign exchange reserves. Figure 3 pegged exchange rate to help bring inflation shows this accumulation as the spike up- down (see Table 1) from 159 percent in wards in foreign exchange reserves at the 1987 to 8.0 percent in 1993. This section ex- beginning of 1994. plains the role of a pegged exchange rate in Finally, a fundamental measure of bringing down inflation and the dangers of whether the real exchange rate is properly such a policy. valued is its effect on exports. The Mexi- To understand how the value of the can government questioned how the real peso affects inflation, consider how mone- exchange rate could be overvalued when tary policy, exchange rates, and prices in- export growth was as strong as it was. Cu- teract. Because only the Bank of Mexico, mulative nonoil export growth from 1985 Mexico’s , can issue peso cur- to 1994 was more than 200 percent, in rency or reserves, within very broad limits, the same range as such export powers as it can control the value of the peso by con- , Korea, Singapore, and .9 trolling the supply of pesos. Similarly, the To summarize: Dornbusch and Werner Bank of Mexico also controls Mexican in- presented evidence that the real exchange flation by increasing or decreasing the rate, as measured by the WPI, was overval- growth of the . No central ued in 1993 and 1994. Although in retro- bank, however, can independently control spect it looks as if Dornbusch and Werner both the exchange rate and inflation at were correct, this was not obvious at the the same time. The desired inflation rate time. Other measures of the exchange rate may not be compatible with the preferred showed no overvaluation, economic re- exchange rate. That is, if a central bank form had likely made historical measures picks a level of inflation to target, it must less reliable than usual, and export growth choose the particular path for the ex- was strong. change rate that is consistent with that inflation rate. By choosing a path for the 9 Data taken from Gil-Diaz and Disinflation and the Overvalued Peso. In exchange rate (and money growth) consis- Carstens (1995). 1993 and 1994 many economists who sup- tent with a low inflation rate, the Bank of 10 See Dornbusch and Werner ported NAFTA warned that the real ex- Mexico could use a pegged exchange rate (1994) and Hufbauer and change rate had become overvalued and as a tool to help lower the inflation rate. Schott (1993).

F EDERAL R ESERVE B ANK OF S T. LOUIS

39 J ULY /AUGUST 1996

There are three ways in which a or band was lowered (devalued) only pegged exchange rate policy helped the slowly. The principle of controlling the ex- fight against inflation. First, a stronger change rate to restrain inflation remained peso forced Mexican producers of tradeable the same, however. goods to restrain price increases to directly As the preceding section concluded, it compete with foreign producers. Second, was not obvious that the peso was over- in every disinflation, the credibility of the valued. To the extent that it may have disinflation is important to breaking the been, however, creating an overvalued ex- momentum of the inflation with little cost change rate by using a pegged exchange in lost output. That is, people have to be rate to bring down inflation is neither convinced that inflation will fall. A pegged new nor unique to Mexico. Numerous au- exchange rate helps break inflationary ex- thors, including Corbo and De Melo pectations by providing a concrete measure (1987), have commented on the tendency of the progress in fighting inflation; it toward overvaluation in the so-called gives the public an inflation-sensitive nomi- “Southern Cone” countries of , nal anchor. People can see that the cur- Chile and when the exchange rency doesn’t free fall against a (low infla- rate is used as an instrument to reduce tion) foreign currency and so they come to inflation. Gil-Diaz and Carstens (1995) believe that inflation is falling. Third, add and Finland to this list of maintaining the exchange rate against the countries that experienced overvaluation. dollar gives the monetary authority instant In all of these countries, there was sub- feedback as to the pressure on the value of stantial real overvaluation but no free the peso. trade agreement to blame for it. The danger with using a pegged ex- change rate to fight inflation is that the real Other Reasons to Avoid Devaluation. Ve- exchange rate will become overvalued if lasco (1995) discusses several reasons why domestic inflation exceeds the rate of depre- the Mexican authorities wished to avoid de- ciation of the domestic currency plus for- valuation in 1994. First, they did not wish to eign inflation. Pegging the nominal ex- lose the gains they had made against infla- change rate while domestic inflation tion. Aside from the domestic consequences exceeds foreign inflation raises the real ex- of loss of control of inflation, the Mexican change rate, and domestic goods become authorities feared that a devaluation would more expensive in terms of foreign goods. be ineffective in changing the relative price This fights domestic inflation for the rea- of Mexican and foreign goods if inflation sons outlined previously, but at the cost would outpace the depreciation of the peso. of making domestic industries less com- Such a devaluation would have been the petitive in tradeable goods. Such a situa- worst of both worlds: more inflation, a loss tion may quickly become unsustain- of credibility and no improvement in the able. competitiveness of domestic goods. Further, Despite this danger, many developing to maintain their credibility with investors, countries with histories of high inflation the Mexican policymakers were reluctant to have used restrictive with devalue even in the face of large shocks. a pegged (or ) exchange rate They were concerned that devaluation to control inflation. That is the course would call into question the policymakers’ Mexico chose; from 1988 to 1994, the commitment to other reforms and result in a Bank of Mexico used the exchange rate as loss of foreign investment. 11 A crawling peg is a pre- an instrument to bring down inflation. announced daily rate of slow Summary on the Value of the Peso. In devaluation. In the target zone The peso was pegged to the dollar in system, Mexican authorities March 1988. In January 1989, the peg was 1993–1994 Dornbusch and Werner pre- pledged to keep the exchange changed to a crawling peg and a moving sented evidence, convincing in retrospect, rate with the dollar within target zone was introduced in December that the peso was overvalued. It was not given margins. 1991.11 The lower limit of the target zone clear at the time, however, that this was the

F EDERAL R ESERVE B ANK OF S T. LOUIS

40 J ULY /AUGUST 1996

case. To the extent the peso may have been Figure 4 overvalued, it was because of the disinfla- tion strategy pursued by Mexico, and other Current Account and Capital Account policy concerns. The evidence is not consis- Balances as a Percentage of GDP tent with the claim that the government of 15 Mexico deliberately manipulated the value Capitol of the peso and planned a devaluation Account 10 years in advance or that the authorities avoided a faster rate of depreciation solely 5 (or primarily) because of the politics of NAFTA. 0 Current Account NAFTA AND -5

INTERNATIONAL -10 CAPITAL FLOWS 1988 1989 1990 1991 1992 1993 1994 1995 This section introduces the concept of Year capital flows, lays out the hypothesis that NAFTA was responsible for the peso crisis by stimulating capital flows out of Mexico, opposite to the capital account balance be- and then shows that the evidence is not cause a country can import more than it consistent with this hypothesis. exports only by selling foreigners claims on existing real or financial assets.12 Thus What Are Capital Flows? a deficit in the current account must be balanced by an equal and opposite capital Capital flows entail the buying and account surplus because the two accounts selling of existing assets. When foreign in- are the opposite sides of the same transac- vestors buy real or financial Mexican as- tion. One measures the net value of the sets, for example, capital flows into Mex- goods and services received, and the other ico. Real assets include factories and real measures the net value of the assets ex- estate; financial assets encompass bonds changed for the goods and services. A na- and equity. Foreign investment is divided tion that runs a current account deficit into foreign direct investment (FDI) and (and, by definition, a capital account sur- portfolio investment. FDI is distinguished plus) is borrowing from abroad, selling as- from portfolio investment by active con- sets like bonds in exchange for new goods trol of the assets: Buying a factory is FDI, and services. A country running a current buying a bond is portfolio investment. account surplus is lending to other coun- The national income accounts measure tries by buying assets in exchange for ex- net capital flows by the balance in the cap- of goods and services. In a world ital account. A surplus in a nation’s capital with balanced trade, there would be no account means that more capital is flowing opportunities for net international borrow- 12 The accumulation or loss of of- into the country than is flowing out; that ing, and domestic savings would have to ficial reserves like foreign ex- is, the country is selling more existing as- equal domestic investment. change, , or other assets sets than it is buying. Similarly, the current Figure 4 illustrates that Mexico ran in- permits an exception to the account measures trade in goods and ser- creasing current account deficits and capi- rule that the current and capital vices, net receipts on foreign investment, tal account surpluses for the period accounts must balance. A na- tion can temporarily finance a and unilateral transfers. A current account 1990–1994. In other words, it was increas- current account deficit by sell- deficit means that a country is importing ingly borrowing from abroad—as much as ing off official assets, as Mex- more newly produced goods and services 8 percent of its GDP by 1993. Capital in- ico did in 1994. This simply than it exports. flows—a capital account surplus—are use- amounts to a change in the Aside from measurement errors, the ful because they permit a nation to con- way that the capital account is current account balance must be equal and sume more and grow faster by borrowing defined.

F EDERAL R ESERVE B ANK OF S T. LOUIS

41 J ULY /AUGUST 1996

against future income. The sustainability 2. NAFTA triggered capital outflows that of capital inflows (borrowing) is limited by led to the peso’s devaluation by creat- the capacity of the borrower to pay back ing political instability. the loan. Borrowing for present consump- tion is not sustainable unless national in- The Case that NAFTA Generated Capital come, or the capacity to pay back the Inflows. The capital inflows to Mexico loan, grows rapidly. Borrowing to invest (Mexico’s capital account surplus) in in productive capacity, borrowing that 1990–1994 meant that Mexico was bor- increases future income or reduces future rowing from abroad to finance its current expenditures is more likely to be sustain- account deficit. A low savings rate made able. Judging whether capital flows are Mexico more dependent on international sustainable is difficult, however, because capital flows and therefore more vulnera- consumption and investment are defined ble to shocks.13 Critics contend that this and measured imperfectly. For example, dependence was critically worsened by spending on , , or passage of NAFTA. There are two ways in consumer durables is counted in the na- which NAFTA might generate capital in- tional accounts as consumption, but per- flows to Mexico. The first is by decreasing haps it should be called investment. Mexican national savings. The second is by increasing the desirability of investment in Mexico. The Case that NAFTA Was Why might NAFTA reduce Mexico’s Responsible for Capital Flows savings rate? First, by directly lowering that Caused the Peso Crisis trade barriers, NAFTA made consumption The immediate precipitating factor in of imports, especially consumer durables, the of December 1994 cheaper and more attractive relative to sav- was the desire of investors to get their as- ing. Given Mexico’s history of protection- sets, especially portfolio investment, out ism, consumers may have feared that free of pesos, which they feared would be trade was temporary and wished to buy devalued, and into dollars or other foreign while they could. A rise in the consump- currency. That is, capital was flowing out tion rate must lower the savings rate be- of Mexico. This section lays out the logic cause all disposable income of a nation or behind the critics’ second hypothesis an individual can be classified as either about NAFTA and the Mexican financial consumption or savings. Second, NAFTA crisis—that NAFTA drove international and other economic reforms may have in- capital flows that led to the devaluation of creased expectations of future income, in- the peso. There are also two versions of creasing Mexicans’ willingness to go into this hypothesis. The first version requires debt and lenders willingness to permit only that NAFTA simply encouraged this.14 At the same time, financial reforms capital inflows—either by depressing na- gave ordinary people greater access to 13 A savings rate and a consump- tional savings or by making Mexico a credit markets and thus greater ability to tion rate are savings and con- go into debt. Finally, if NAFTA con- sumption, respectively, as per- more attractive investment environment— centages of income. and that capital inflows, in the form of tributed to an artificially higher real value portfolio investment, are inherently dan- of the peso, that would have also made 14 The Permanent Income Hypoth- gerous. The second version suggests that imported goods much less expensive and esis, developed by economist NAFTA generated political instability consumption more attractive. (1957), pre- dicts that people base their that sparked capital outflows and the consumption on their lifetime devaluation. These hypotheses require “... NAFTA served as a kind of income. That is, they smooth that: ‘Good Housekeeping Seal of Ap- their consumption over time by proval’ that encouraged even more borrowing during periods of 1. Either NAFTA encouraged international investors into Mexico.” low income and saving during capital inflows, which are intrinsically Anderson, Cavanagh and Ranney periods of high income. destabilizing, or (1996), p. 3.

F EDERAL R ESERVE B ANK OF S T. LOUIS

42 J ULY /AUGUST 1996

The second form of the capital inflow tween Mexico’s prosperous north hypothesis suggests that NAFTA may have and an impoverished south.” generated capital inflows to Mexico by mak- , January 8, 1994. ing Mexico a more attractive investment en- vironment. This hypothesis would explain The uprising was soon contained by the surge, in early 1994, of capital inflows the , but it and other political that caused the peso to appreciate. NAFTA shocks concerned investors throughout the was considered especially important to in- year. They engendered fears that the eco- vestors because an international treaty nomic reforms in Mexico had moved too made the reforms more likely to be perma- fast and would lead to social unrest that nent. There is considerable reluctance would roll back the reforms. In fact, the ini- to break a treaty with a foreign govern- tial devaluation on December 20, 1994, was ment. sparked by a run on the peso started by ru- An implicit assumption of the hypothe- mors of renewed fighting in .15 sis that NAFTA was responsible for the peso These political shocks led investors to crisis because it encouraged capital inflows exchange pesos for dollars at the Bank of is that such flows are inherently destabiliz- Mexico, causing a series of falls in Mexico’s ing. Portfolio investment, in particular, was foreign exchange reserves, limiting its 15 See Gil-Diaz and Carstens frequently maligned as being a cause of the short-term ability to defend the peso.16 Fig- (1995) or IMF (1995) for the crisis. It was said to be moved on a whim ure 3 illustrates the stepwise falls in foreign details of the decision to de- with a short-term investment horizon, cre- exchange reserves during 1994. By the time value. ating financial market volatility. Such a view that rumors of renewed fighting rattled the 16 In the long run, the Bank of requires that international capital markets markets on December 19, 1994, the Bank Mexico used its control over be subject to fads or speculative bubbles. of Mexico had nearly run out of foreign ex- the money supply to determine Critics point to the volatility of the dollar in change reserves. Without foreign exchange the foreign exchange value of the 1980s, the European Exchange Rate to defend the peso, the Bank of Mexico had the peso. Over the short term, Mechanism crises of 1992 and 1993, and the to devalue.17 Critics of NAFTA might argue however, the Bank of Mexico recent flood of capital into emerging mar- that the treaty caused the peso crisis by defended the value of the peso kets as evidence of this. sparking the Chiapas uprising. by buying and selling pesos for dollars. By itself, this action The Case that NAFTA Contributed to Cap- would reduce the supply of ital Outflows Through Political Instability. Evaluating the Evidence on NAFTA pesos and push up Mexican in- terest rates. The Bank of Mex- From the Mexican view, the purpose of and Capital Flows ico, however, fully sterilized NAFTA was to create a more prosperous This section evaluates the evidence on the purchase of pesos by buy- and stable Mexico. Nevertheless, even good NAFTA and capital flows to see whether it ing outstanding bonds in ex- economic policy can unintentionally create is consistent with either of the hypotheses change for pesos, putting the dislocations and political instability. Some that NAFTA caused the peso crisis through pesos back into circulation. have charged that NAFTA contributed to its effect on capital flows. The first subsec- Sterilization is intended to the Chiapas uprising that triggered the tion examines the evidence on the extent leave domestic interest rates capital outflows that brought down the to which NAFTA encouraged capital in- unchanged after foreign ex- change purchases or sales. peso. flows and the next looks at the argument that capital flows are inherently destabiliz- 17 Some suggest that the Bank of “On January 1, 1994—the day ing. Finally, the role of NAFTA in the Chi- Mexico could have used its that the North American Free- apas uprising and political instability is ap- control over the domestic Trade Agreement (NAFTA) took praised. money supply to defend the effect, binding Mexico’s moderniz- peso, but it was reluctant to do this because of the effect high ing economy to that of the United Evidence on NAFTA and Capital Inflows. interest rates would have had States—Indian peasants at the Mexico did indeed have low and falling na- on the real economy and the southern end of the country rose tional savings rates—4 percent of GDP in banking sector. Certainly by in armed rebellion. ... Many in 1994, for example (see Table 1)—making it December 1994 this strategy Chiapas fear that NAFTA will more dependent on international capital would have imposed large worsen the existing divide be- flows. Net savings fell from 10.8 percent of costs.

F EDERAL R ESERVE B ANK OF S T. LOUIS

43 J ULY /AUGUST 1996

GDP in 1990 to 4.1 percent of GDP in Part of the capital inflow was soaked up 1994.18 This reduction in savings was dri- in the form of a rapid increase and then ven by corresponding increases in private decrease in official reserves—shown by and government consumption, which rose the spike in Figure 3 at the beginning 2.5 percentage points and 4.1 percentage of 1994. That is, the Bank of Mexico points, respectively, over the same period. bought up dollars in sterilized intervention There are several problems with the to keep the price of the peso down in hypothesis that the declining savings was a January and February 1994. The surge result of NAFTA. First, it is not very plau- was not out of proportion to earlier flows, sible that NAFTA would cause a large rise however. in private (or government) consumption. To the extent that private Mexican Trade barriers cause consumers to substi- consumption increased in the early 1990s, tute one form or source of consumption there are many factors aside from NAFTA for another but change aggregate levels of to explain it. Prolonged slow growth consumption/saving relatively little. Thus (since 1980) had created repressed con- the effect of trade liberalization on trade sumer demand. After growth returned in deficits is not likely to be very big. Also, 1988, consumption spending rose along the fact that most of the increase in con- with it. Also, to the limited extent that re- sumption was caused by a rise in govern- ducing trade barriers may change savings ment consumption does not fit well with and consumption decisions, NAFTA was the hypothesis that NAFTA caused the fall not the only trade initiative. Mexico en- in savings. The sluggish economy in 1993 gaged in unilateral trade liberalization and and election year politics in 1994 were trade agreements with Chile, , more likely than NAFTA to have played a , and . Similar to role in this relaxation of fiscal policy. Fi- other developing countries, economic re- nally, the timing of the inflows is wrong; form and financial liberalization—quite the inflows started in 1990 with the reso- apart from NAFTA—raised expectations of lution of the and the liberaliza- increased future income and gave more tion of capital account rules to permit for- Mexicans access to credit. eigners to hold government bonds and To summarize: the evidence does nonvoting equity shares in Mexican not support the argument that NAFTA firms.19 Figure 4 illustrates this rise in the drove large capital inflows to Mexico. capital account surplus; the majority of NAFTA did increase foreign confidence capital inflows had entered Mexico well and marginally increased capital inflows, before NAFTA was negotiated, much less but most capital inflows had entered before enacted. Other economic reforms, like the passage of NAFTA. In fact, NAFTA may decline in inflation and the of have delayed a crisis by drawing in capital state-owned industries, also helped drive that supported the peso in early 1994. investment. While NAFTA was not the only or Volatility of Capital Flows. The question even the major causal factor for capital of whether capital flows are excessively inflows, it probably had some marginal volatile or inherently destabilizing is diffi- effect. Figure 4 shows that Mexico’s capi- cult to answer because capital should exit tal account surplus did peak in the first a country in response to poor economic quarter of 1994, coinciding with the im- policies or other factors that reduce its plementation of NAFTA. The figure is productivity. This helps ensure that capital somewhat deceptive in that a surge in is as productive as possible and provides inflows in January 1994 and February governments with an immediate incentive 18 Data from OECD (1995) Na- 1994 was masked in the quarterly capital to maintain sound policies. On the other tional Accounts. account figures by a major outflow in hand, it is possible that portfolio invest- 19 See Sachs, Tornell and Velasco March after the assassination of presiden- ment overreacts to information, and this (1995a and 1995b). tial candidate Luis Donaldo Colosio. volatility does create problems.

F EDERAL R ESERVE B ANK OF S T. LOUIS

44 J ULY /AUGUST 1996

Because capital does move rapidly out capital flows provide major advantages for of weak in moments of crisis both investors and recipients. and these movements can be destabilizing, This movement of assets can also some economists have advocated a very cause difficulties, however. Corbo and small tax on international financial trans- Hernandez (1996) studied the problems actions to deter short-term .20 posed by this movement of assets in nine Trying to eliminate international capital countries: Argentina, Chile, Columbia, In- flows would be a mistake, however, be- donesia, Malaysia, Mexico, the Philip- cause capital inflows can be quite helpful pines, , and Thailand. They re- in promoting development. Also, they are port that though the absolute level of not necessarily destabilizing. Rather, their investment in Mexico from 1986 to 1993 volatility can depend on the soundness of was very large compared with the other macroeconomic policies followed in the countries, Thailand, Malaysia, and Chile recipient countries. Further, outflows received larger capital inflows as a percent- occur without regard to the nationality of age of GDP than did Mexico. Many of the investors in the presence of unsound these countries have also encountered macroeconomic policies. Domestic resi- problems similar to those confronted by dents would get their money out of the the Mexican authorities. For example, in domestic assets under the same conditions regimes with fixed or predetermined ex- as international investors, if the value of change rates, capital inflows can lower do- these assets were threatened. mestic interest rates, raise domestic expen- ditures and temporarily raise inflation, NAFTA and the Chiapas Uprising. which can lead to an overvalued currency NAFTA may have been a catalyst for, but and large trade deficits. certainly was not the cause of, the Chiapas Partly to offset the tendency toward uprising. This rebellion reflected griev- overvaluation caused by the capital flows, ances long and deeply felt by the impover- all of these countries have undertaken lib- ished south against the more prosperous eralization of trade, though none of them north. Also, the uprising was only one has concluded a trade agreement compara- political shock among many that Mexico ble in importance to NAFTA. But free endured that year, including two major trade agreements are not necessary to cre- assassinations, a rise in U.S. interest rates ate substantial capital inflows. The breadth and a presidential election. If the Decem- and size of these capital flows to reforming ber Chiapas uprising had not sparked the countries in the developing world in the crisis, something else likely would have. last 10 years makes it difficult to believe that NAFTA was the primary reason for Capital Flows to Emerging Markets. Mex- the inflows to Mexico. ico is not the only to Capital flow volatility poses particular experience heavy capital inflows recently. problems for fixed exchange rate regimes In the last 10 years capital inflows to de- because capital outflows are synonymous veloping countries have increased sharply with exchange rate crises. Investors who because of two factors: market-oriented perceive a possibility of a discrete fall in policy reforms and low interest rates pre- the value of their assets (that is, a devalu- vailing in the developed world. These fac- ation), will attempt to get their money out tors draw in capital because policy reforms of the weak currency. Thus crises appear raise the return to investment in develop- suddenly when capital is easily moved. ing countries and the low interest rates in These outflows are merely a symptom of the developed world provide a less attrac- the problem, however, not the cause. tive alternative for international investors. For developing countries, capital flows Summary of the Evidence on NAFTA and provide a much needed source of funds for Capital Flows. NAFTA is an unlikely cul- 20 See Frankel (1996) for a short . Ideally, international prit to blame for the quantity of capital in- discussion of the .

F EDERAL R ESERVE B ANK OF S T. LOUIS

45 J ULY /AUGUST 1996

flows Mexico received in the early 1990s. REFERENCES The surge in capital flows started well be- Aguilar, Linda M. “NAFTA: A Review of the Issues,” Federal Reserve fore the enactment of NAFTA and had Bank of Chicago Economic Perspectives (Jan/Feb 1993), more to do with the rise in consumption pp. 12–20. by the Mexican government and the other economic reforms undertaken. Anderson, Sarah, John Cavanagh and David Ranney, eds. NAFTA’s First Whatever the source or timing of the Two Years—The Myths and the Realities. Washington D.C.: The Insti- inflows, however, NAFTA was not respon- tute for Policy Studies, 1996. sible for the outflows. It is generally ac- Brown, Drusilla K., Alan V. Deardorf and Robert M. Stern. “A North knowledged that the outflows were gener- American Free Trade Agreement: Analytical Issues and a Computa- ated by some combination of inconsistent tional Assessment,” The World Economy (January 1992), policies and political shocks that generated pp. 11–29. a ; the Mexican government Calvo, Guillermo. “Capital Flows and Macroeconomic Management: had more short-term obligations—in the Tequila Lessons,” unpublished manuscript (March 1996). form of dollar-linked bonds—coming due Corbo, Vittorio, and Jaime de Melo. “Lessons from the Southern Cone 21 than it had liquid assets. Policy Reforms,” Research Observer (July 1987), pp. 111–42. CONCLUSION ______, and Leonardo Hernandez. “Macroeconomic Adjustment to As Mexico entered into NAFTA at the Capital Inflows: Lessons from Recent Latin American and East Asian beginning of 1994, it was widely and cor- Experience.” World Bank Research Observer (February 1996), pp. 61–85. rectly applauded as a model of economic re- form. Before the end of the year, however, it Dornbusch, Rudiger, and Alejandro Werner. “Mexico: Stabilization, Re- was forced to first devalue and later to allow form and No Growth,” Brookings Papers on Economic Activity, vol. 1, the peso to float. In early 1995, it was forced (1994), pp. 253–97. to borrow from the IMF and the United Freidman, Milton. “A Theory of the Consumption Function.” Princeton States to get through a liquidity crisis. University Press, 1957. Critics of NAFTA such as Ross Perot, Garber, Peter M., and Subir Lall. “The Role and Operation of Derivative 21 Because the Bank of Mexico Pat Buchanan, William Greider, and Markets in Crises,” unpublished manuscript had ultimate control over the Robert Kuttner blamed the trade treaty (February 1996). supply of pesos, it is true, by for this crisis. This article examines two Gil-Diaz, Francisco, and Agustin Carstens. “One Year of Solitude: Some definition, that the devaluation versions of this argument: that Mexican was caused by insufficiently Pilgrim Tales About Mexico’s 1994-1995 Crisis,” The American policymakers manipulated the value of Economic Review (May 1996), pp. 164–9. tight monetary policy. That is, the peso because of NAFTA and that the Mexican authorities’ ex- NAFTA caused volatile international capi- ______, and ______. “Some Hypotheses Related to the Mexi- change rate and growth policy tal flows that brought down the peso. can 1994-1995 Crisis,” Banco de Mexico, Serie Documentos de In- objectives were mutually incon- vestigacion (1995) 9601. sistent. For a discussion of the The evidence does not support the hy- policy priorities of the Mexican pothesis that the crisis could have re- Greider, William. “Southern Comfort. How Come There Are Billions to authorities, see Gruben (1996) sulted from NAFTA’s economic effects. Bail Out Mexico But Nada for U.S. Cities?” Rolling Stone (March 9, or Gil-Diaz and Carstens Any peso overvaluation in 1994 resulted 1995), pp. 40–2. (1995). Overviews of the from the use of the exchange rate to re- Gruben, William C. “Policy Priorities and the Mexican Exchange Rate Cri- events leading to the crisis may duce inflation, a common consequence of sis,” Federal Reserve Bank of Dallas Economic Review (First Quarter be found in OECD (1995), IMF this strategy. Although capital inflows can 1996), pp. 19–29. (1995) or GAO (1996). present problems and aggravate instability Calvo (1996) and Garber and Hufbauer, Gary Clyde, and Jeffrey J. Schott. NAFTA: An Assessment. In- in developing countries, they are also stitute for International Economics, 1993. Lall (1996) discuss the roles very useful to promote economic devel- played by capital flows and opment. In any case, the flows to Mexico International Monetary Fund. “Evolution of the Mexican Crisis,” in Inter- weakness in the Mexican finan- were only partially driven by NAFTA. national Capital Markets. Developments, Prospects, and Policy Issues, cial system. Sachs, Tornell and IMF (1995), pp. 53–69. Velasco (1995a and 1995b) NAFTA was not, in any sense, responsi- discuss the problems created ble for the devaluation, but this episode Kehoe, Timothy J. “A Review of Mexico’s Trade Policy from 1982 to by Mexico’s dollar-linked debt reminds us that good policies can have 1994,” The World Economy: Global Trade Policy (1995), pp. instruments. unintended consequences. 135–51.

F EDERAL R ESERVE B ANK OF S T. LOUIS

46 J ULY /AUGUST 1996

Krugman, Paul. “Review of: NAFTA: An assessment,” Journal of Eco- nomic Literature (June 1995), pp. 849–51. ______. “The Uncomfortable Truth about NAFTA: Its Foreign Policy, Stupid,” Foreign Affairs (Nov.-Dec. 1993), pp. 13–19. Kuttner, Robert. “Trouble in Mexico no surprise, misguided NAFTA to drop in peso,” Akron Beacon Journal (January 22, 1995), p. C1. “Mexico’s second-class citizens say enough is enough,” The Economist (January 8, 1994), pp. 41–3. Organization for Economic Co-operation and Development. OECD Eco- nomic Surveys 1994-1995, Mexico (1995). Oliver, Christian. Conference Summary of “Did NAFTA Kill the Peso?” American Enterprise Institute for Public Policy Research, January 30, 1996. Orme, William. “Myths versus Facts: The Whole Truth about the Half- Truths,” Foreign Affairs (Nov.-Dec. 1993), pp. 2–12. Perot, Ross. “Perspective on NAFTA,” Los Angeles Times (January 4, 1995), p. 7. Rowen, Hobart. “Administration Ignored Peso Warnings,” The Washing- ton Post (February 5, 1995), p. H2. Sachs, Jeffrey, Aaron Tornell and Andres Velasco. “The Collapse of the Mexican Peso: What Have We Learned?” National Bureau of Eco- nomic Research Working Paper (June 1995a), 5142. ______, ______and ______. “The Real Story,” The Inter- national Economy (March/April 1995b), pp. 14–17 and 50–1. Schott, Jeffrey J. “NAFTA: An American Perspective,” International Trade Journal (Spring 1994), pp. 3–8. Tornell, Aaron, and Gerardo Esquivel. “The Political ’s Entry to NAFTA,” National Bureau of Economic Research Working Paper (October 1995), 5322. United States General Accounting Office (GAO). “Mexico’s Financial Cri- sis. Origins, Awareness, Assistance, and Initial Efforts to Recover.” Re- port to the Chairman, Committee on Banking and , House of Representatives, February 1996. United States Senate. “The Mexican Peso Crisis,” Hearings before the Committee on Banking, Housing, and Urban Affairs (Jan. 31, 1995, March 9-10, May 24, 1995, and July 14, 1995). Velasco, Andres. “Lessons from the Recent Mexican Crisis,” CV Starr Newsletter (vol. 13, 1995).

F EDERAL R ESERVE B ANK OF S T. LOUIS

47