This Selected Issues paper analyzes reserve adequacy in Mexico. Reserve adequacy has been of renewed interest, as the authorities have introduced a new rules-based mechanism of U.S. dollar sales to reduce the rate of reserve accumulation. The paper examines the recent experience with Mexico’s fiscal management tools in light of the need for further consolidation. It reviews evidence on the cyclical behavior of fiscal policy in Mexico, finding that it has generally been procyclical since the early 1990s. The paper also examines the determinants of inflation in Mexico since 1997.

Abstract

This Selected Issues paper analyzes reserve adequacy in Mexico. Reserve adequacy has been of renewed interest, as the authorities have introduced a new rules-based mechanism of U.S. dollar sales to reduce the rate of reserve accumulation. The paper examines the recent experience with Mexico’s fiscal management tools in light of the need for further consolidation. It reviews evidence on the cyclical behavior of fiscal policy in Mexico, finding that it has generally been procyclical since the early 1990s. The paper also examines the determinants of inflation in Mexico since 1997.

I. Reserve Adequacy in Mexico1

This chapter analyzes reserve adequacy in Mexico. Reserve adequacy has been of renewed interest, as the authorities have introduced a new rules-based mechanism of dollar sales to reduce the rate of reserve accumulation. The strong growth in Mexico’s international reserves since the Tequila crisis has led to a current level that seems adequate according to various criteria, including debt amortizations in the coming year, the risk of capital flight, and country-specific risks. Finally, projections of future reserve accumulation indicate that reserve adequacy is not likely to be at risk under baseline assumptions, even though the new mechanism would slow accumulation relative to the previous mechanism. Under an alternative scenario, with lower oil prices and less external financing of government operations, vulnerability ratios could start declining, pointing to the possible need for a more flexible mechanism over time to ensure reserves remain adequate.

A. Introduction

1. Mexico has built up substantial international reserves in recent years, strengthening investor confidence and improving access to international capital markets for both the public and private sectors. Recent experience with financial crises has shown that both holding and managing sufficient reserves, and disclosing adequate information to market participants, helps countries to prevent and better cope with external crises.2

2. Mexico’s reserve accumulation during 1996–2001 was helped by a purchase mechanism of foreign receipts, despite a sale mechanism of reserves to moderate exchange rate volatility. The authorities put in place two mechanisms between 1996 and 2001, with different objectives. The first one (dollar put options) was aimed at accumulating reserves, as commercial banks bid for the right to sell U.S. dollars to the central bank during times of peso strength. The central bank accumulated close to US$12 billion through this mechanism. The second mechanism was aimed at moderating exchange rate volatility, with the central bank selling dollars to the market in times of significant declines in the value of the peso. This mechanism became operational ten times, with cumulative sales of around US$2 billion. Both mechanisms were suspended in 2001. Subsequently, the central bank moved to a system of no sales of dollar receipts, and reserves rose from US$45 billion at end-2001 to US$55 billion by mid-2003.

3. Analysis of reserve adequacy for Mexico has been of renewed interest, as the authorities have introduced a new rules-based mechanism of daily dollar sales to reduce the rate of reserve accumulation. The authorities argue that, given the current high level of international reserves and the anticipated pace of accumulation in the immediate future, the cost of holding reserves is progressively yielding lower net benefits. As a result, they announced a new rules-based mechanism to reduce the rate of reserve accumulation of reserves in March 2003. It specifies that 50 percent of net international reserves (NIR) accumulated in the previous quarter (excluding sales of reserves under the mechanism) would be sold in daily auctions in the next quarter, as long as the net quarterly reserve accumulation would surpass US$250 million (net of the dollar sales undertaken through this mechanism).3 The mechanism is intended to reduce the rate of foreign reserves accumulation according to a rule, not as an intervention device to stabilize the exchange rate. In this regard, the authorities execute the daily sales at a pre-set time, eliminating the possibility of auctioning dollars at moments where the peso could be under pressure.

4. The main findings of the paper are as follows:

  • Strong growth in Mexico’s international reserves since the Tequila crisis has led to a current level that seems adequate according to various adequacy ratios. Reserves currently surpass two benchmarks for adequacy that take into account debt amortizations in the coming year, the risk of capital flight, and country-specific risks.

  • Projections on gross international reserves indicate that reserve adequacy is not likely to be at risk under baseline assumptions, even though the new mechanism would slow reserve accumulation relative to the previous mechanism.

  • Under an alternative scenario over the medium term, with lower oil prices and less external financing devoted to reserve accumulation, adequacy ratios would start declining. In this context, an alternative mechanism could be called for to ensure reserve adequacy is maintained.

5. The structure of the paper is as follows. Section B compares the level of reserves in Mexico with those in other emerging economies with sustained access to international capital markets using various indicators. Reserve adequacy is then evaluated according to two simple benchmarks. Section C discusses alternative projections of reserve accumulation. Section D concludes.

B. Are Current Reserves Adequate?

6. The literature on reserve adequacy, while highlighting the need for an appropriate level of reserves, has not provided firm guidance as to what specific level this should be. Hence, in assessing reserve adequacy, it is necessary to rely on a variety of indicators, complemented by empirical studies on the revealed demand for reserves.4 The analysis here considers four different indicators. First, the ratio of reserves to imports is considered, although recent studies note its declining importance. In particular, external vulnerability is no longer defined merely by current account shocks, but also by capital account shocks, as evidenced the Asian financial crisis. Second, the analysis looks at two vulnerability indicators: the ratio of reserves to short-term debt on a remaining maturity basis (STD); and the ratio of reserves to broad money. The first assesses a country’s risk when it loses access to international capital markets, while the second measures the potential impact of a loss of confidence in the domestic currency, leading to capital flight. Third, the study looks at the ratio of reserves to dollarized deposits, which may be relevant given that the central bank cannot create dollar liquidity to offset a run on such deposits. Finally, reserves are viewed relative to net public sector external debt, indicating the capacity of the government to meet its external financing requirements without recourse to new sources of funds.

7. Mexico has been accumulating international reserves at a strong pace since the Tequila crisis, leading to an improvement in all of the above adequacy ratios (Figure 1). This trend is in line with that observed in other emerging market economies. By end-June 2003, reserves5 stood at US$55.3 billion, covering nearly four months of imports and 116 percent of short-term debt on a remaining maturity basis. This contrasts with the weak reserve position during the Tequila crisis; when reserves fell to US$4.9 billion, or 0.7 months of imports, in January 1995.

Figure 1.
Figure 1.

Mexico: International Reserves and Vulnerability Ratios

Citation: IMF Staff Country Reports 2004, 250; 10.5089/9781451955613.002.A001

Source: Data from BOM, SHCP, IFS, and IMF staff estimates, except for data on short-term debt on a remaining maturity basis, which is from BIS/OECD/IMF/World Bank (series G, H, I), and IMF staff estimates.

8. Although Mexico complies with the crude traditional rule of thumb of holding reserves covering three months of imports, it compares less favorably in this respect with some other emerging market economies. Column 3 in Table 1 reports monthly average import coverage in 2002 for the countries in our sample.6 The overall sample displays an average coverage of 8.6 months of imports. Only South Africa and Hungary had lower ratios than Mexico. This low coverage for imports is explained by the fact that maquiladora-related imports (in-bond assembly for re-export) are included. Excluding maquiladora imports (which accounted for 35 percent of gross imports in 2002) would lead to an increase in import coverage to around six months.

Table 1.

Mexico: Reserve Adequacy Indicators for Twenty Emerging Market Economies

article image

Flexible exchange rate refers to managed floating or independently floating exchange rates regimes. Fixed exchange rate regime refers to conventional fixed peg arrangements or pegged exchange rates within horizontal bands. For more details, see Annual Report on Exchange Arrangements and Exchange Restrictions (IMF, 2003).

In weeks of imports.

In percent. Ratios on reserves-to-imports and on reserves-to-broad money are monthly averages for 2002. The ratio of reserves-to-short-term-debt (on a remaining maturity basis) refers to 2002Q2. Data Sources: All data are from the International Financial Statistics (IMF), except for short-term debt on a remaining maturity basis, which is from the Joint BIS/IMF/OECD/World Bank Statistics (series G (liabilities to banks due within a year), H (debt securities issued abroad due within a year), and I (non-bank trade credits due within a year)). Data on reserves refer to non-gold reserves (line 1.1.d). Broad money is computed as the sum of money (line 34) and quasi-money (line 35).

9. As noted above, capital account considerations seem more relevant in light of recent international experience. In this regard, Mexico’s reserves seem sufficient to withstand well a capital account crisis due to a reversal in confidence. When analyzing the coverage of broad money in terms of reserves, Mexico compares well with the overall sample, with reserves covering around 40 percent of broad money, slightly above the sample average of 37 percent. Within Latin America, however, Mexico shows one of the lowest ratios, second after Brazil (29 percent). Chile, Colombia, and Peru hold reserves covering above 50 percent of broad money, and Venezuela holds reserves well above that level (60 percent). Mexico has a relatively low degree of dollarized deposits, particularly in relation to the rest of the region (at about 10 percent of M1).7 It also operates a fully flexible exchange rate regime. The level of reserves appears, in this respect, to be adequate when weighted by the risk of a capital flight in Mexico. Mexico’s attractiveness to international and domestic investors is among the strongest in the Western Hemisphere, and the country is one of only two countries in Latin America benefiting from an investment-grade rating.

10. Mexico complies well with the recommendation that reserves should cover external debt amortization for a year, in the event of a halt in international market access. In this regard, the ratio of reserves to short-term external debt on a remaining maturity basis (STD) is considered the most important indicator for emerging market economies. Several early warning system (EWS) studies have found that low levels of reserves with respect to STD have increased the probability of crisis.8 In this regard, a rule of thumb has been put forward suggesting that reserves should allow a country to forego access to capital markets during a year, by covering short-term debt on a remaining maturity basis.9 Column 5 in Table 1 reports the ratio of reserves to STD.10 Mexico has a ratio of 132 percent in end-June 2002, although the ratio compares less favorably with respect to the sample average (233 percent). This is mainly due to the surge in reserve accumulation in the Asian economies, to an average above 400 percent of STD in mid-2002.11 Within Latin America, Mexico’s reserves are higher than the average of 125 percent.

11. The level of reserves covers almost two thirds of net public sector external debt (NPED). This ratio could be interpreted as the ability of the country to pay its net liabilities without having to rely on any borrowing and regardless of the amortization profile. This ratio confirms the trend in reserve accumulation and shows the efforts of the Mexican authorities to reduce vulnerabilities in the economy to a capital account shock. At end-June 2003, reserves covered around 70 percent of net public sector external debt, compared with 6 percent at the time of the Tequila crisis in January 1995.

12. Mexico’s reserves are currently above a composite minimum benchmark estimated taking into account external debt amortizations, the risk of a capital flight, and country-specific risks.12 In particular, Mexico’s reserves are compared with the minimum benchmark for holdings computed according to a methodology suggested by De Beaufort and Kapteyn (2001). Starting from the minimum of full coverage of short-term external debt,13 an estimate of the potential for capital outflow stemming from residents is added to the minimum holding of reserves. A percentage of broad money is used as a proxy for potential capital flight. According to this measure, the risk of capital flight is indicated by 5 and 10 percent of broad money for countries like Mexico that have a flexible exchange rate.14 Finally, the fraction of broad money covered is adjusted using an index of country risk. In this regard, the benchmark considers the international country risk index (CRI) for Mexico, which covers political, financial, and economic risks.15 While reserves were somewhat below the composite benchmark until 2002, the level as of 2003 is 15 percent above the estimated adequacy range (Table 2).

Table 2.

Mexico: Computation of a Minimum Reserve Benchmark

(In billions of U.S. dollars, unless specified)

article image

On a remaining maturity basis.

The benchmark interval covers short-term debt plus 5 (10) percent of broad money for the lower (upper) limit, adjusted by the country risk factor.

Sources: Data on short-term debt are from BOM and Fund staff estimates. Data on broad money are from IFS (IMF) and Fund staff estimates. Data on reserves from BOM and Fund staff estimates. Country risk index (CRI) is from the international country risk guide (ICRG).

13. The level of reserves is also adequate when the risk of capital flight is assessed against dollarized deposits. The minimum benchmark is computed by including 100 percent of dollarized deposits, instead of the coverage of 5 and 10 percent of broad money.16 The analysis indicates that reserves were below the benchmark until 2001. Following the subsequent increase to US$45 billion, reserves rose to 26 percent above the suggested upper limit of the benchmark by end-2002.

14. The benefits of having international reserves need to be balanced against the holding costs. According to the literature on the demand for reserves, a country balances the macroeconomic adjustment costs incurred if reserves are exhausted against the opportunity costs of holding reserves. As international reserves are usually held in the form of short-term interest bearing assets, the opportunity costs of holding reserves will be the difference between the domestic return on capital and the return obtained from holding the reserves. When analyzing the costs and benefits of holding reserves, positive externalities associated with higher reserve accumulation, including for example lower costs of external borrowing by both the public and private sectors, should also be taken into account.

C. Projections of Reserve Accumulation

15. A baseline projection for reserve accumulation would be consistent with a strengthening of the reserve position, although at a slower pace than in recent years (Figure 2). Accumulation depends mainly on oil export receipts and government financing operations in international markets (including PEMEX). Positive prospects for external financing opportunities and for continued strong oil exports (despite the baseline assumption of a decline in world oil prices, to around US$21.5 dollar per barrel over the medium term) would translate into further increases in Mexico’s international reserve position, and thus further dollar sales by the central bank. Given current projections for oil prices and external financing estimates, Mexico would continue accumulating reserves strongly, with the level reaching US$75.2 billion by 2008. Despite the fact that new mechanism would reduce the rate of reserve accumulation, reserves are projected to increase by at least US$3.5 billion per year from 2004 onwards. By comparison, during 2002, Mexico accumulated reserves of US$5.5 billion per year.

Figure 2.
Figure 2.

Mexico: Projections of Reserve Accumulation

Citation: IMF Staff Country Reports 2004, 250; 10.5089/9781451955613.002.A001

Source: BOM and IMF staff estimates

16. Under this scenario, both the ratio of reserves to short-term debt and to gross external financing requirements would continue to rise (Figure 2). In order to maintain these ratios constant, reserves would need to grow on average by 2.5 percent annually throughout 2008. In contrast, projected reserves rise by nearly 4 percent annually. The ratio of reserves to short-term debt (on a remaining maturity basis) would increase to nearly 160 percent, and reserves would cover 110 percent of the financing requirements of the balance of payments by 2008, up from 143 percent and 103 percent respectively in 2003.

17. Thus, under the baseline scenario, reserves would well surpass the minimum benchmarks over the medium term. Assuming a constant country risk index, reserve projections would be 22 percent, on average, above the upper limit of the minimum benchmark defined by De Beaufort and Kapteyn (2001), considering broad money coverage from 2003 onwards (Table 2). When considering full coverage of dollarized deposits, reserves would again be above the minimum benchmark during the projection period, by an average margin of 57 percent.

18. Under an alternative scenario with significantly lower oil prices and less external financing of government operations (including PEMEX), reserve adequacy ratios could decline over the medium term. This alternative scenario is constructed assuming a reduction of 50 percent of PIDIREGAS project financing going to international reserves accumulation, and oil prices for the Mexican mix falling sharply to US$10 per barrel over the medium term. In this case, the ratio of reserves to short-term external debt on a remaining maturity basis would decline by 2 percent over the medium term. The ratio of reserves to external financing requirements would decline significantly to 87 percent in 2008. Under this scenario, although reserve projections would exceed the minimum benchmark defined by De Beaufort and Kapteyn (2001), the margin would decline to only 6 percent in 2008.

D. Conclusions

19. Reserves currently appear to be adequate in Mexico. Different vulnerability ratios indicate that reserves compare well with levels in other emerging market economies. They comply with the rule of covering short-term debt on a remaining maturity basis, the most important vulnerability ratio. Reserves would also be sufficient to face a capital flight crisis related to a loss in the confidence of the domestic currency by residents, even though such an event has a low possibility of occurrence.

20. Although the baseline projection for reserves shows a continuing strengthening in the reserve position, lower oil prices and less external financing could lead to declining adequacy ratios. Reserves are projected to accumulate strongly despite the implementation of the new mechanism, by around US$3.5 billion per year over the medium term. Under the baseline scenario, adequacy ratios would continue their upward trend, and reserves would well surpass the reserve level suggested by De Beaufort and Kapteyn. Under an alternative scenario, with lower oil prices and less external financing of government operations, adequacy ratios could start declining, suggesting the possible need for an alternative mechanism to ensure that reserve adequacy is maintained.

References

  • Berg, A., E. Borensztein, G. Milesi-Ferreti, and C. Patillo, 1999, “Anticipating Balance of Payment Crisis: The Role of Early Warning Systems”, Occasional Paper No.186, International Monetary Fund, Washington D.C.

    • Search Google Scholar
    • Export Citation
  • Brassière, M. and C. Mulder, 1999, “External Vulnerability in Emerging Market Economics: How High Liquidity Can Offset Weak Fundamentals and the Effects of Contagion”, IMF Working Paper, WP/99/88, International Monetary Fund, Washington D.C.

    • Search Google Scholar
    • Export Citation
  • De Beaufort Wijnholds, J. and A. Kapteyn, 2001, “Reserve Adequacy in Emerging Market Economies”, IMF Working Paper, WP/01/143, International Monetary Fund, Washington D.C.

    • Search Google Scholar
    • Export Citation
  • Economist Intelligence Unit, 2003, Country Risk Service, Mexico Report, London.

  • Edison, H., 2003, “Are Foreign Exchange Reserves in Asia too High?”, World Economic Outlook, forthcoming, International Monetary Fund, Washington D.C. (October).

    • Search Google Scholar
    • Export Citation
  • Edwards, S., 1985, “On the Interest-Rate Elasticity of the Demand for International Reserves: Some Evidence from Developing Countries”, Journal of International Money and Finance, 4, 287-95.

    • Search Google Scholar
    • Export Citation
  • Faal, E. and N. Thacker, 2003, “Dollarization in Latin America: Is Mexico Different?”, Mimeograph, International Monetary Fund, Washington D.C.

    • Search Google Scholar
    • Export Citation
  • Greenspan, A., 1999, “Currency Reserves and Debt”, remarks before the World Bank, Conference on Trends in Reserves Management, Washington D.C.

    • Search Google Scholar
    • Export Citation
  • International Country Risk Guide, Political Risk Services, IBC, USA Publications.

  • International Monetary Fund, 2001, “Issues in Reserve Adequacy and Management”, Country Report No. 02/237, Washington D.C.

  • International Monetary Fund, 2002, “Assessing Fiscal and External Sustainability: The Case of Mexico”, in Country Report No. 02/238, 99-108, Washington D.C.

    • Search Google Scholar
    • Export Citation
  • Tweedie, A., 2002, “The Demand for International Reserves—A Review of the Literature”, Note 2 in External Review of Quota Formulas—Annex, International Monetary Fund, Washington D.C.

    • Search Google Scholar
    • Export Citation
1

Prepared by M. Vera-Martin.

2

See IMF (2001) for a more general discussion on reserve adequacy and management.

3

First implemented in May, the authorities started with daily auctions of US$32 million for the 3-month period from May to July. The Bank of Mexico (BOM) would sell US$14 million per day during the second 3-month period the system is operational, i.e. between August and October. Initial estimates suggested that total sales in 2003 would be US$3–4 billion.

4

IMF (2001) recommends complementing the analysis of reserve adequacy indicators with stress testing of the balance of payments to allow for a better understanding of the interaction between reserve adequacy, vulnerability, and country-specific factors and policies. An analysis of external sustainability for Mexico with stress testing of the balance of payments was prepared during the 2002 Article IV Consultation (see Country Report No. 02/238, IMF (2002)).

5

The figures on reserves refer to the series “Total Reserves minus Gold”, which includes the U.S. dollar value of monetary authorities’ holdings of SDRs, reserve position in the Fund, and foreign exchange. This definition is different from the concept of “Net International Reserves”, due to the fact that the latter excludes short-term liabilities with the Federal Government, Pemex, and other creditors, and includes liabilities with the IMF.

6

The analysis considers 20 large emerging market economies that have generally enjoyed more or less uninterrupted access to international capital markets. The sample accounts for all emerging market regions (Asia, Eastern Europe, and Latin America). Data for the cross-country analysis is described in Table 1.

9

Guidotti was apparently the first to propose such a rule at a seminar of the Group of 33 in Bonn in the spring of 1999. Greenspan (1999) also put forward such a rule, complemented by two enhancements—that average maturity of external debt should be above a certain threshold, and that countries implement a liquidity-at-risk standard.

10

These data on short-term external debt on a remaining maturity basis are collected from creditor sources, and may differ from the data reported in individual IMF staff reports, which are usually obtained from the authorities.

11

See Edison (2003) for a discussion on excess reserve accumulation.

12

These results are in line with those reported in Edison (2003), which show that Mexico accounts for the bulk of reserve accumulation in Latin America since 1997, and has actual reserves above those warranted by fundamentals.

13

As the analysis here is no longer cross-country comparative, data on short-term debt on a remaining maturity basis are from the Bank of Mexico and Fund staff projections, in line with the figures reported on Table 2 of the Staff Report for the 2003 Article IV Consultation.

14

The authors suggest to adjust by a fraction of between 10 and 20 percent of M2 for managed float and fixed exchange rate regimes, and between 5 and 10 percent for floaters. This interval is consistent with the declining trend of the standard deviation of the reserves-to-broad money ratio, which ranges between 8 percent to 6 percent depending on the period considered.

15

The Economist’s country risk index was also considered, yielding similar conclusions. Benchmark intervals are reported on the basis of the CRI, as it implied a more conservative level of reserve adequacy.

Mexico: Selected Issues
Author: International Monetary Fund