Abstract
This Selected Issues paper analyzes fiscal multipliers in Mexico. Estimates of fiscal multipliers--obtained from state-level spending--fall within 0.6-0.7 after accounting for dynamic effects. However, the size of multipliers varies with the output gap. The planned fiscal consolidation-under the estimated multipliers-is projected to subtract on average 0.5 percentage points from growth over 2015-20. However, there are offsetting effects. The positive growth impulse of lower costs on manufactured goods production is estimated to reach 0.5 percentage point in 2015 and 2016, largely offsetting the impact of fiscal consolidation on growth in the near term.
Trade and Financial Spillovers to Mexico1
Economic activity in emerging markets has slowed down in recent years, which has been partly attributed to less favorable external conditions. For Mexico, close linkages with the U.S. has benefited Mexico over the years, but also exposes the country to fluctuations in U.S. economic activity and changes in U.S. monetary conditions. Moreover, the open capital account and large foreign holdings of Mexican assets expose Mexico also to financial spillovers, including the possibility of abrupt changes in capital flows. A Structural Bayesian VAR analysis for the period of 2001Q3–2015Q2 suggest that while increases in U.S. growth are transmitted over one-for-one to Mexico (a cumulative impact of 1.22 percent after 8 quarters), an increase in emerging market risk premiums are transmitted less than one-for-one (a cumulative impact of -0.65 percent after 8 quarters). The results suggest that Mexico could outgrow a negative shift in market sentiment toward emerging markets if accompanied by a pickup in US growth. Moreover, by using a decomposition of the U.S. long-term interest rates we find that a rise in U.S. bond yields that is not accompanied by higher U.S. growth has a negative effect on Mexico’s output.
A. Introduction
1. Mexico has close trade and financial ties with the global economy, and especially with the United States. The U.S. is by far the largest recipient of Mexico’s manufacturing and agricultural exports, and is also the main source of portfolio and foreign direct investment flows to Mexico, possibly explaining the close correlation of the business cycles of the two economies. Export growth has been a key driver of economic activity in Mexico, with significant spillovers to domestic demand. Focusing on financial linkages, international investors held about half of total government debt in mid-2015 (including 36 percent of local-currency government bonds), as well as a large share of corporate bond debt. Mexico’s gross portfolio investment liabilities account for 37 percent of GDP, of which 25 percent of GDP are debt securities.
U.S. and Mexico Real GDP Growth
(In percent)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: National Authorities; and IMF staff calculationsU.S. and Mexico Real GDP Growth
(In percent)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: National Authorities; and IMF staff calculationsU.S. and Mexico Real GDP Growth
(In percent)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: National Authorities; and IMF staff calculations
Contribution to Real GDP Growth
(In percent)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: National authorities; Haver Analytics; and IMF staff calculations.Contribution to Real GDP Growth
(In percent)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: National authorities; Haver Analytics; and IMF staff calculations.Contribution to Real GDP Growth
(In percent)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: National authorities; Haver Analytics; and IMF staff calculations.2. This study quantifies the spillover effects of external shocks on Mexico’s economic activity. The next two sections describe in more detail the trade and financial linkages between Mexico and the United States. After that, a Bayesian vector autoregression is estimated to evaluate the impact of external shocks (real and financial) on Mexico’s GDP and other domestic variables. Finally, the paper uses a decomposition of the drivers of changes in U.S. long-term interest rates into real, monetary, and risk shocks, and looks at the differential effects of these shocks on Mexico’s growth and long-term interest rates.
B. Trade Linkages
3. Since the signing of the North American Free Trade Agreement (NAFTA) in 1994, Mexican exports have increased nearly fivefold. The automobile industry and the telecommunication sector accounted for a large part of the expansion. Mexico’s share of world exports also increased somewhat after NAFTA, and has stabilized around 2 percent more recently. The U.S. remains Mexico’s main trading partner, accounting for 80.5 percent of total exports. Canada accounts for 2.7 percent, and the next three largest partners, Spain, China, and Brazil account for about 1½ percent each.
Total Exports of Goods
(USD, billion)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: UNCOMTRADE; and IMF staff calculations.Total Exports of Goods
(USD, billion)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: UNCOMTRADE; and IMF staff calculations.Total Exports of Goods
(USD, billion)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: UNCOMTRADE; and IMF staff calculations.
Total Exports by Destination, 2014
(In percent of total)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Source: Direction of Trade Statistics; and IMF staff calculations.Total Exports by Destination, 2014
(In percent of total)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Source: Direction of Trade Statistics; and IMF staff calculations.Total Exports by Destination, 2014
(In percent of total)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Source: Direction of Trade Statistics; and IMF staff calculations.
Total Imports by Origin, 2014
(In percent of total)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Total Imports by Origin, 2014
(In percent of total)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Total Imports by Origin, 2014
(In percent of total)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
4. Mexico’s total share in the U.S. market has been stable at around 10 percent. There has been a temporary decline in the early to mid-2000s, when China started gaining market share in the U.S. after entering the World Trade Organization (Chiquiar, Fragoso and Ramos-Francia, 2008). However, Mexico has regained market share in recent years, partly due to changes in relative labor costs in favor of Mexico.
U.S. Trading Partners
(Share of U.S. imports)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: UNCOMTRADE and IMF staff calculations.U.S. Trading Partners
(Share of U.S. imports)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: UNCOMTRADE and IMF staff calculations.U.S. Trading Partners
(Share of U.S. imports)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: UNCOMTRADE and IMF staff calculations.
Unit Labor Cost
(Index, 2008=100)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: National authorities; Haver Analytics; and IMF staff calculations.Unit Labor Cost
(Index, 2008=100)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: National authorities; Haver Analytics; and IMF staff calculations.Unit Labor Cost
(Index, 2008=100)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: National authorities; Haver Analytics; and IMF staff calculations.5. Industrial production has been very highly correlated with the United States, reflecting Mexico’s integration in the North American production value chains. The detailed breakdown of sectors suggests strong integration in manufacturing production, and especially in chemicals and machinery. Bank of Mexico (2015) shows high correlation between Mexican exports and U.S. industrial production. Moreover, the analysis shows that Mexican and American industrial production are cointegrated and follow a common cycle.
C. Financial Linkages
Mexico and U.S. Industrial Production
Mexico and U.S. Industrial Production
| Sector | Correlation | ||
|---|---|---|---|
| Total | 0.94 | ||
| Mining | −0.25 | ||
| Utilities | 0.89 | ||
| Electric power | 0.90 | ||
| Manufacturing | 0.93 | ||
| Paper | −0.85 | ||
| Printing | −0.38 | ||
| Petroleum and coal | 0.66 | ||
| Chemical | 0.88 | ||
| Plastics and rubber | 0.30 | ||
| Nonmetallic minerals | 0.01 | ||
| Primary metal | 0.47 | ||
| Fabricated metals | 0.64 | ||
| Machinery | 0.73 | ||
| Computer and electronics | 0.26 | ||
| Electrical equipment and appliances | 0.00 | ||
| Furniture | 0.34 | ||
Mexico and U.S. Industrial Production
| Sector | Correlation | ||
|---|---|---|---|
| Total | 0.94 | ||
| Mining | −0.25 | ||
| Utilities | 0.89 | ||
| Electric power | 0.90 | ||
| Manufacturing | 0.93 | ||
| Paper | −0.85 | ||
| Printing | −0.38 | ||
| Petroleum and coal | 0.66 | ||
| Chemical | 0.88 | ||
| Plastics and rubber | 0.30 | ||
| Nonmetallic minerals | 0.01 | ||
| Primary metal | 0.47 | ||
| Fabricated metals | 0.64 | ||
| Machinery | 0.73 | ||
| Computer and electronics | 0.26 | ||
| Electrical equipment and appliances | 0.00 | ||
| Furniture | 0.34 | ||
6. Foreign portfolio liabilities have risen steadily over the years. The greatest increase in foreign holding occurred in domestic peso-denominated bonds, of which nonresidents now hold about 11 percent of GDP (USD 144 billion). There has also been an increase in private sector bonds issued abroad from 2.5 percent of GDP in 2002 to 5.8 percent of GDP (USD 75 billion) in 2014. To some extent this reflects a substitution of external loans by bond debt, as total private sector debt rose more slowly from 8.7 percent of GDP in 2002 to 10.5 percent in 2014. Nonresident holdings of Mexican equity also doubled since 2002, reaching almost 12 percent of GDP in 2014. About half of foreign portfolio and direct investment liabilities are held by U.S. investors.
Nonresidents’ Holdings of Portfolio Liabilities
(In percent of GDP)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: National authorities; and IMF staff calculations.Nonresidents’ Holdings of Portfolio Liabilities
(In percent of GDP)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: National authorities; and IMF staff calculations.Nonresidents’ Holdings of Portfolio Liabilities
(In percent of GDP)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: National authorities; and IMF staff calculations.
Foreign Liabilities
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: Coordinated Portfolio Investment Survey (2014); and Coordinated Direct Investment Survey (2013).Foreign Liabilities
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: Coordinated Portfolio Investment Survey (2014); and Coordinated Direct Investment Survey (2013).Foreign Liabilities
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: Coordinated Portfolio Investment Survey (2014); and Coordinated Direct Investment Survey (2013).7. Mexican long-term interest rates and spread dynamics are influenced significantly by external developments. Long-term yields on 10-year domestic-currency government bonds in Mexico have fallen from 11 ½ in 2001 to around 6 percent by mid-2013. During this period, U.S. yields also dropped from around 5 ½ percent to 3 ½ percent.2 Nonetheless, since mid-2013, while U.S. yields continued to drop, Mexican yields rose, in part as global attitudes to risk, as approximated by EMBI global spreads, affected also Mexico’s risk spreads.
Bond Yields and Spreads
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: Bloomberg L.P.; Haver Analytics; and IMF staff calculations.Bond Yields and Spreads
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: Bloomberg L.P.; Haver Analytics; and IMF staff calculations.Bond Yields and Spreads
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Sources: Bloomberg L.P.; Haver Analytics; and IMF staff calculations.8. Mexican interest rates react strongly to changes in U.S. rates.
The 2014 Article IV Report on Mexico presented results from a vector error correction model, suggesting that changes in long-term U.S. interest rates transmit more than one-to-one to Mexican domestic sovereign bond yields. A 100-basis point shock to 10-year U.S. rates boosts Mexican yields by 140 basis points. A variance decomposition analysis shows that 50 percent of fluctuations in Mexico’s 10-year yields are explained by innovations in U.S. rates.
The IMF’s 2015 Spillover Report shows more generally that U.S. (and euro area) monetary developments affect bond yields elsewhere. In case of pure monetary shocks, a 100-basis point increase in U.S. interest rates leads to a rise by about 30 basis points in emerging markets and non-systemic advanced economies interest rate after three months, combined with a small capital outflow. An interest rate increase that occurs as a result of stronger U.S. growth would boost yields in emerging markets and non-systemic advanced economies by about 40 basis points. In this case, however, it would be accompanied by capital inflows.
The IMF’s 2014 Regional Economic Outlook for the Western Hemisphere finds that on average increases in U.S. rate boost yields by about half as much, although spillovers can be much larger than one-to-one during exceptional times. This occurred, for example, following the taper tantrum episode, which occurred when interest rate where at all-time lows in many countries and suddenly snapped up. In the specific case of Mexico, yields rose by about 200 basis points, and more than half of the increase appears to be the result of a reaction to the U.S. monetary shock. Moreover, the report shows using a full-fledged macro model, that a positive U.S. growth shock would have a positive impact on Mexican GDP, even if there is a simultaneous increase in emerging market risk premiums.
Ebeke and Kyobe (2015) consider the impact of U.S. rates on emerging markets depending on the participation of foreign investors in local government bond markets. They find that if foreign holdings exceed 35 percent of the market—as is the case in Mexico—the transmission of a 100-basis point shock to U.S. rates is amplified and results in a rise in emerging market yields of 140 basis points (compared to just 40 basis points for countries where foreign ownership is below the 35 percent threshold).
D. The Impact of Foreign and Domestic Factors on GDP
9. A structural BVAR is estimated to quantify the size of growth spillovers and the impact on various transmission channels (trade and financial). Following the approach used in WEO April 2014 (Chapter 4), this allows an estimation of the domestic and external factors in explaining output growth. The Bayesian VAR approach is particularly suited, because it allows the handling of a large number of variables and lags in a relatively small sample. The sample covers the quarters from 2001Q3 to 2015Q2.
10. External variables aim to capture key external shocks to the Mexican economy while domestic variables aim to capture the transmission mechanisms. External variables include the U.S real GDP growth, the 10-year U.S Treasury bond rate, and the composite emerging market economy bond spread (J.P Morgan Emerging Market Bond Index (EMBI)). As domestic variables, we consider Mexican real GDP growth, real exports, the rate of change of the real exchange rate against the U.S. dollar, ten-year local-currency sovereign bond spreads and gross portfolio inflows as a share of GDP. The inclusion of domestic ten-year spreads constrains the starting point of the analysis as Mexican Bonos were first issued in July 2001.
11. The SBVAR is identified using the Cholesky decomposition. The ordering of the endogenous variables yt follow:
The first three variables constitute the external block and the remaining variables the domestic block. The external variables are assumed to not respond to the internal variables contemporaneously. All variables enter the model with four lags.3
12. A historical decomposition of real GDP suggests that external factors explained more than 2/3 of the sharp dip in Mexico’s growth during the global financial crisis. Given the estimates from the reduced-from VAR, growth can be expressed as the sum of initial conditions and all the structural shocks in the model. The sum of the shocks from the identified external factors provides the contribution of all external factors.4 A historical decomposition of growth shows how external factors played a large role in explaining the Mexican real GDP growth in the past years (explaining 70 percent of the sharp dip in Mexico’s growth during the global financial crisis). The historical decomposition shows the deviation of Mexico’s growth from average where the average is computed as quarterly growth for the period of 2001Q3–2015Q2 (2.5 percent). The increase in the contribution of a factor is measured by the change in its level relative to the previous quarter.
Historical Decomposition: Mexico Real GDP Growth
(Deviations from average, in percent)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Source: IMF staff calculations.Historical Decomposition: Mexico Real GDP Growth
(Deviations from average, in percent)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Source: IMF staff calculations.Historical Decomposition: Mexico Real GDP Growth
(Deviations from average, in percent)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Source: IMF staff calculations.13. The role of internal factors has increased over 2013–14. The historical decomposition of real GDP growth also suggests that external factors do not explain the slowdown in domestic growth in recent years (explaining only 20 percent of the overall negative contribution to growth in 2013Q1–2015Q2).5 This finding suggests that the role of internal factors could have risen in the past two years.6 Interestingly, this phenomenon seems to be common to other large emerging markets (e.g., Brazil, China, India) where external factors could not account for the slowdown in growth in the recent years (WEO, 1014). Very recently, the negative contribution of external factors to growth started to increase. Moreover, in recent months, a stabilization of oil production has been observed and the construction sector began to recover since the second half of 2014. Thus, in 2015, a faster growth of domestic demand is registered.
14. Increases in U.S. growth are transmitted over one-for-one to Mexico. On impact, a 1 percentage point shock in U.S. growth (equivalent to 1.8 standard deviations) raises Mexican growth by 0.5 percentage point. The cumulative effects remain positive beyond the short term and reach 1.22 after 8 quarters. The result is somewhat in line with the average impact of world growth on Latin America (Osterholm and Zettelmeyer, 2007). Despite a different dynamic, the cumulative impact of an increase in U.S. growth on Mexico is similar to the response of U.S. growth with respect to its own shock, also reflecting the effect of the propagation of the shock through the U.S. economy.
15. Meanwhile, increases in emerging market’s risk premia are transmitted less than one-for-one. A 100-basis point increase in the composite EMBI spread (equivalent to 2.2 standard deviations) reduces Mexico’s growth by 0.62 on impact. The cumulative effects remain negative beyond the short term reaching -0.65 after 8 quarters. The analysis suggest that with a pickup in U.S. growth, Mexico could outgrow the impact of “some” negative shift in market sentiment toward emerging markets.
Impact of External Shock on Domestic Growth
(Percentage points)
Impact of External Shock on Domestic Growth
(Percentage points)
| One Percentage Point Shock | |||
|---|---|---|---|
| Response | U.S. Real GDP Growth (1 STD=0.56) | EMBI Spread (1 STD=0.46) | |
| Domestic Real GDP | On Impact | 0.50 | −0.62 |
| Average growth (yoy), quarters 1-4 | 0.94 | −0.78 | |
| Average growth (yoy), quarters 5-8 | 0.28 | 0.13 | |
| Cumulative impact after 8 quarters | 1.22 | −0.65 | |
Impact of External Shock on Domestic Growth
(Percentage points)
| One Percentage Point Shock | |||
|---|---|---|---|
| Response | U.S. Real GDP Growth (1 STD=0.56) | EMBI Spread (1 STD=0.46) | |
| Domestic Real GDP | On Impact | 0.50 | −0.62 |
| Average growth (yoy), quarters 1-4 | 0.94 | −0.78 | |
| Average growth (yoy), quarters 5-8 | 0.28 | 0.13 | |
| Cumulative impact after 8 quarters | 1.22 | −0.65 | |
E. A Closer Look at Changes in Monetary Conditions in the U.S.
16. U.S. long-term interest rates reflect the expected path of U.S. monetary policy, expected U.S. economic performance, and general risk attitudes. Each of these components may affect Mexico differently; therefore, we consider a decomposition of the U.S. 10-year bond yield into real, monetary, and risk shocks following Buitron and Vesperoni (2015), which allows the identification of these shocks controlling for changes in risk-appetite.7
17. The role of money shocks in explaining U.S. 10-year bond yield, which peaked after the taper talk, faded by end-2014. The decomposition highlights the relative contribution of money shocks in the aftermath of the taper talk in May 2013. Prospects of a better economic outlook seem to have played a greater role by end-2013 during the subsequent taper announcement (see WEO, 2014).
1. U.S. 10-Year Yield
(percent)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Source: Buitron and Vesperoni (2015).1. U.S. 10-Year Yield
(percent)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Source: Buitron and Vesperoni (2015).1. U.S. 10-Year Yield
(percent)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Source: Buitron and Vesperoni (2015).18. We assess the differential impact of money and real shocks in economic activity and financial variables. The model described above is estimated now considering the following external variables—real shocks, money shocks, VIX—while keeping the same set of domestic variables.8
19. U.S. monetary policy shocks spill over into Mexican yields and initially dampen Mexican growth, while U.S. real shocks are positive for Mexican growth and have a smaller impact on yields. Figure 1 shows the responses of a 100-bps increase in the U.S. bond yield, with the red (blue) line displaying the impact of real (money) shocks. Unexpected stronger economic prospects in the U.S. associated with an increase in the U.S. yield leads to higher spreads and improved economic activity. The shocks also boost investors’ risk-appetite, which causes capital to flow to Mexico and the currency to appreciate in real terms. Money shocks are followed by higher domestic bond spreads in Mexico, a (small) real depreciation of the currency, a decrease in gross portfolio inflows, and lower real GDP growth.
Mexico: Response to a 100bps Shock in 10-Year U.S. Bond Yield
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Source: IMF staff calculations.Mexico: Response to a 100bps Shock in 10-Year U.S. Bond Yield
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Source: IMF staff calculations.Mexico: Response to a 100bps Shock in 10-Year U.S. Bond Yield
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Source: IMF staff calculations.F. Conclusions
20. External factors played an important role in driving the drop in growth during the global financial crises but less so in recent years. Shocks idiosyncratic to Mexico during 2013–14, such as the contraction of construction activity in 2013, and lower oil production, are possible explanations. Looking ahead, Mexico remains well placed to gain from a pickup in U.S. growth.
References
Arora, V. and M. Cerisola, 2001, “How does U.S. Monetary Policy Influences Sovereign Spreads in Emerging Markets?” IMF Staff Papers 45(3): 474–98.
Bank of Mexico, 2015, Sincronización de la Producción Manufacturera Mexicana con la de Estados Unidos, Informe Trimestral, Abril-Junio 2015, Banco de México, pp. 24–25.
Buitron, C. and E. Vesperoni, 2015, “Spillover Implications of Differences in Monetary Conditions in the United States and Euro Area,” International Monetary Fund, Washington, D.C.
Chiquiar, D., E. Fragoso and M. Ramos-Francia, 2007, “La Ventaja Comparativa y el Desempeño de las Exportaciones Manufactureras Mexicanas en el Periodo 1996–2005,” Banco de México Working Paper No. 2007–12.
Ebeke, C. and A. Kyobe, 2015, “Global Financial Spillovers to Emerging Market Sovereign Bond Markets”, IMF Working Papers, No. 15/141.
International Monetary Fund (IMF), 2014a, 2014 Spillover Report, IMF Policy Papers (Washington).
International Monetary Fund (IMF), 2014b, Rising Challenges, Regional Economic Outlook: Western Hemisphere (Washington).
International Monetary Fund (IMF), 2014c, Mexico: 2014 Article IV Consultation—Staff Report, IMF Country Report No. 14/319 (Washington).
International Monetary Fund (IMF), 2014d, World Economic Outlook (Washington, April).
International Monetary Fund (IMF), 2015, 2015 Spillover Report, IMF Policy Papers (Washington).
Kamil, H. and J. Zook, 2012, “What Explains Mexico’s Recovery of U.S. Market Share?,” Selected Issues Paper, IMF Country Report No. 12/317.
Litterman, R., 1986, “Forecasting with Bayesian Vector Autoregressions: Five Years of Experience,” Journal of Business and Economic Statistics, Vol. 4, No. 1, pp. 25–38.
Matheson, T. and E. Stavrev, 2014, “News and Monetary Shocks at a High Frequency: A Simple Approach,” Economics Letters, Vol. 125(2), pp. 282–286.
Osterholm, P. and J. Zettelmeyer, 2007, “The Effects of External Conditions on Growth in Latin America,” IMF Working Paper, WP/07/176, 2007.
Sims, C., and T. Zha, 1998, “Bayesian Methods for Dynamic Multivariate Models,” International Economic Review, Vol. 39, No. 4, pp. 949–68.
Uribe, M., and V. Z. Yue, 2006, “Country Spreads and Emerging Countries: Who Drives Whom?” Journal of International Economics 69(1): 6–36.
Appendix I. Historical Decomposition of Real GDP Growth (Includes U.S. Industrial Production)
Historical Decomposition: Mexico Real GDP Growth
(Deviations from average, in percent)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Source: IMF staff calculations.Historical Decomposition: Mexico Real GDP Growth
(Deviations from average, in percent)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Source: IMF staff calculations.Historical Decomposition: Mexico Real GDP Growth
(Deviations from average, in percent)
Citation: IMF Staff Country Reports 2015, 314; 10.5089/9781513593432.002.A002
Source: IMF staff calculations.Appendix II. Impulse Responses (Section D)
Appendix III. Impulse Responses (Section E)
Prepared by Juliana D. Araujo and Alexander Klemm. The authors thank Dora Iakova and Fabian Valencia for valuable comments and suggestions and Alexander Herman for outstanding research assistance.
While higher U.S. rates have coincided with wider spreads on foreign-currency debt in EMs (Arora and Cerisola (2001), Uribe and Yue (2006)), increased foreign investment in local bond markets could have contributed to greater linkages between U.S. and domestic yields in Mexico (see next paragraph for a discussion).
Modified “Minnesota priors” (Litterman, 1986) were used where each variable is assumed to follow a first-order autoregressive process with independent, normally distributed errors and coefficients of 0.8. The relative weight of the prior, as in Sims and Zha (1998), was set to
The remaining shocks are treated as residual and are termed internal. As such, the residual shocks could also partly embody other factors such as common or exogenous shocks (e.g., natural disasters).
By conducting a similar exercise but including U.S. and Mexico Industrial Production instead of real GDP growth, we might better capture trade linkages across both economies. The exercise yields similar qualitative results but it suggests a lower positive contribution of external factors in the last 5 quarters of the sample period (see Appendix I).
Some Mexico specific factors that could have helped explain the deceleration in activity, which are expected to revert going forward, include the reduction of oil production in 2013 and 2014 as well as the decrease in housing construction due to financial problems in some housing development companies.
Here we focus on the impact of changes in monetary conditions in the U.S. although euro area shocks are also computed. In a first stage, stock prices and bond yields in the U.S. are stripped out from risk-appetite shocks measured by the VIX. In a second stage, real and money shocks within the U.S. are identified as in Matheson and Stavrev (2014).
We convert the identified monthly shocks to quarterly frequency by summing the shocks within each quarter. The sample period is 2001Q3-2014Q4.
