Abstract
Mexico is an open economy with strong real and financial links to the rest of the world with risks of spillovers from global turbulence. Recent gains in market share in the U.S. manufacturing market are owed to improved relative unit labor costs and reemergence of a location advantage. Mexico’s current fiscal framework requires measures to offset the emerging challenges of a decline in oil revenues and the projected increase in health- and pensions-related spending. The sustained increase of bank credit after the global crisis has been reversed. The effects of migration depend on labor reform.
IV. Mexico: Financial Sector Development1
While Mexico’s bond and equity markets have developed rapidly during recent years, banking sector credit has remained relatively subdued compared to emerging market peers. Evidence suggests that bond and equity financing in Mexico can be explained by usual factors (economic development, saving rate, sovereign debt market depth, institutional/regulatory environment), but bank financing has remained below what those factors would suggest. The sustained increase of bank credit after the global crisis has helped to start reversing this puzzle, with potential implications for access to credit for small and medium size enterprises and growth.
A. Background
1. There is growing consensus in the literature that financial development matters for economic growth. Since the early work by Diamond (1969) on the relationship between financial development and growth, substantial work has been done, particularly during the last 20 years.2 Financial development can have an impact on economic growth through different channels, including by (i) facilitating diversification of risks; (ii) mobilizing savings; (iii) reducing the cost of external financing for firms; and (iv) disseminating information and monitoring investment decisions. The empirical literature also suggests that financial depth is important for growth.3
2. The structure of the financial system also matters, as it may affect access to credit for some companies, in particular small and medium size enterprises (SMEs). Although the literature has not found a direct link between the relative development of bank-based and market-based financing and economic growth, it is generally acknowledged that bond and equity financing is less effective in serving borrowers in situations characterized by “opaqueness” (i.e., the difficulty to ascertain whether firms have the capacity or the willingness to pay). Banks, through a number of transactional technologies, are better suited to serve companies for which the lender has to rely predominantly on soft information.4
3. The development of domestic bond and equity markets has helped Mexico recover financing levels reached during the early 1990s. Total credit to the private sector (including bank credit as well as bond and equity financing), which peaked at more than 50 percent of GDP in 1994 (Figure 1), fell to less than 40 percent in the early 2000s and has only recently recovered past high levels (data sources for Figures 1–4 included in Appendix Table 1). This recovery has been mainly driven by the gradual expansion of bond and equity markets between 1994 and 2010 (Figures 2a and 2b), which has likely been driven in part by the reform of the pension system.5 Domestic bond markets, in particular, have grown significantly and reached market capitalization levels of around 20 percent of GDP, larger than the average in other EMs. Equity markets have also developed in line with other EMs, although equity financing is still lower than in comparator countries (this financing includes the cumulative value of annual Initial Public Offerings, as percent of GDP in the year of issuance).6 Despite the development of market financing, total financing in Mexico still appears low compared to other EMs, including other Latin American economies.
Total Private Sector Financing
(share of GDP)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
Total Private Sector Financing
(share of GDP)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
Total Private Sector Financing
(share of GDP)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
Domestic Private Bonds
(market capitalization, share of GDP)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
Domestic Private Bonds
(market capitalization, share of GDP)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
Domestic Private Bonds
(market capitalization, share of GDP)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
Equity
(equity financing, share of GDP)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
Equity
(equity financing, share of GDP)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
Equity
(equity financing, share of GDP)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
4. Bank credit has remained relatively subdued (Figure 3). After reaching almost 40 percent of GDP in 1994, bank credit to the private sector (defined as claims to the private sector by other depositary corporations, as reported in IFS) hit a trough at about 13 percent of GDP in 2004. By 2010, bank financing had rebounded to about 20 percent of GDP, still well below previous peaks.7 Bank credit to GDP also represents less than half the level of other EMs in the sample. More recently, bank credit has shown a sustained recovery, growing at about 10 percent in real terms during the last couple of years.
Bank Credit to Private Sector
(share of GDP)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
Bank Credit to Private Sector
(share of GDP)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
Bank Credit to Private Sector
(share of GDP)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
5. As a result of these trends, the financial sector in Mexico has become less bank-based, compared both to the past and to other EMs (Figure 4). The share of bank credit in total financing to the private sector, which reached almost 75 percent in 1994, was about 35 percent in 2010.8 This contrasts with the experience of other EMs, in which the share of bank credit has remained at around 50–60 percent during the last 20 years.
Banking Credit
(share of total private financing)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
Banking Credit
(share of total private financing)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
Banking Credit
(share of total private financing)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
B. Financial Development: Mexico in the Context of Emerging Markets
6. To study financial sector development in Mexico a cross-country econometric analysis was performed on the drivers of bank- and market-based financing between 1990s and 2010 (Appendix 1). The main results show that:
The evolution of market-based financing in Mexico is explained by usual fundamentals, in line with other EMs. Regression analysis suggests that the dynamics of market financing is driven by the size of the economy, saving rates, the development of sovereign debt markets, institutional factors, foreign financing spreads, and the cost of equity financing. Moreover, market-based financing is usually higher after financial crises, with Mexico’s well captured by the regression analysis without a significant fixed effect.
The evolution of bank credit in Mexico, however, remains only partly explained by usual fundamentals. The analysis suggests that the factors explaining market-based financing help also understand bank credit in EMs. However, while these variables help explain market-based financing in Mexico, the specification for bank financing cannot reject a significant negative fixed effect, suggesting that that macroeconomic and financial factors would predict higher bank credit in Mexico.9
A sequential analysis of crises suggests that the protracted stagnation of bank credit in Mexico is not likely associated with the direct impact from the 1994 crisis. Figure 5 (appendix) shows that a crisis dummy for Mexico becomes significant only in 2005, a decade after the 1994 financial crisis, suggesting that other factors not associated with the immediate financial distress may be at play in the evolution of bank credit, including a lower level of savings through bank deposits and the implementation of tighter prudential regulations that usually follow financial crises.10
Appendix 1. Cross-Country Analysis of Factors Behind Financial System Development
This appendix analyzes the factors behind the evolution of bank- and market-based financing in EMs. Panel regressions comprising the EMs in the sample are used for the period 1990–2010.11 The dependent variables are: (i) the ratio of financing instruments issued in capital markets (i.e., bonds and equity) to GDP; (ii) the ratio of bank financing to the private sector to GDP; and (iii) the ratio of financing instruments issued in capital markets to total financing to the private sector.12 The method of estimation is GMM, including country dummies.13
Three specifications are presented. An initial specification regressing the dependent variables on development, macroeconomic and financial factors is followed by an analysis of the impact of financial crises. Finally, institutional factors are added to the specification.
Basic Specification
Bank credit and market-based instruments are affected inter alia by the level of development, savings, the depth of sovereign debt markets and funding costs (Table 2). The results in the basic specification suggest that: (i) both bank-based and market-based financing increase with the size of the economy; (ii) increasing levels of savings (i.e., total savings for market-based instruments and bank deposits for bank credit) are associated with higher financing; (iii) the development of sovereign debt markets helps boosting market-based instruments, but it has a negative impact on bank financing; (iv) countries with more open capital accounts, by facilitating access of domestic corporations to external credit markets, show lower levels of domestic market-based financing; (v) higher international interest rates and more expensive domestic equity markets are associated with lower market-based financing in EMs; and (vi) higher international interest rates are associated with more bank financing, which is likely funded to a large extent by domestic deposits.
Summary of Variables
Summary of Variables
| Variables | Source(s) | Units |
|---|---|---|
| GDP per capita | WEO, Sept. 2011 | USD |
| National savings as percent of GDP | WEO, Sept. 2011 | Percent |
| Government domestic bonds market capitalization as percent of GDP | BIS, Quarterly Statistics Table 16B WEO, Sept. 2011 | Percent |
| Private domestic bonds market capitalization as percent of GDP | BIS, Quarterly Statistics Table 16A WEO, Sept. 2011 | Percent |
| Capital Account openness | Chinn Ito Financial Openness Index (normalized) | Between 0 and 1 |
| Deposits as percent of GDP | IFS | Percent |
| Bank credit | IFS | Percent |
| Equity market financing as percent of GDP | S&P Global Stock Markets Factbook, 2011 Dealogic | Percent |
| Equity markets index | S&P Global Stock Markets Factbook, 2011 | Index, 2000=100 |
| U.S. 2-Years Treasury Bond | Haver Analytics | Percent |
| World Bank Governance Indicator | World Bank | Percentile |
Summary of Variables
| Variables | Source(s) | Units |
|---|---|---|
| GDP per capita | WEO, Sept. 2011 | USD |
| National savings as percent of GDP | WEO, Sept. 2011 | Percent |
| Government domestic bonds market capitalization as percent of GDP | BIS, Quarterly Statistics Table 16B WEO, Sept. 2011 | Percent |
| Private domestic bonds market capitalization as percent of GDP | BIS, Quarterly Statistics Table 16A WEO, Sept. 2011 | Percent |
| Capital Account openness | Chinn Ito Financial Openness Index (normalized) | Between 0 and 1 |
| Deposits as percent of GDP | IFS | Percent |
| Bank credit | IFS | Percent |
| Equity market financing as percent of GDP | S&P Global Stock Markets Factbook, 2011 Dealogic | Percent |
| Equity markets index | S&P Global Stock Markets Factbook, 2011 | Index, 2000=100 |
| U.S. 2-Years Treasury Bond | Haver Analytics | Percent |
| World Bank Governance Indicator | World Bank | Percentile |
| Variable | Variables used to construct, and source(s), if not defined above | Definition |
|---|---|---|
| Capital markets as percent of GDP | pvtbond sm_adjusted Nominal GDP (in USD): WEO, 2011 | |
| Capital markets as percent of total private financing | Pvtbond sm adjusted creditpvt |
| Variable | Variables used to construct, and source(s), if not defined above | Definition |
|---|---|---|
| Capital markets as percent of GDP | pvtbond sm_adjusted Nominal GDP (in USD): WEO, 2011 | |
| Capital markets as percent of total private financing | Pvtbond sm adjusted creditpvt |
Bank and Market-Based Financing: Basic Specification
Bank and Market-Based Financing: Basic Specification
| (1) | (2) | (3) | |
|---|---|---|---|
| Market-based financing as share of GDP | Banking financing as share of GDP | Market-based as share of total financing | |
| GDP per capita in USD | 0.0022*** | 0.0024*** | 0.0000 |
| Domestic Savings | 0.8958*** | 0.8313** | |
| Public Bond Market | 0.2735*** | -0.5146*** | 0.4197*** |
| Capitalization | |||
| Capital Account | -10.9993** | -4.4179 | 0.4574 |
| Openness | |||
| T-Bond, 2 YRS | -1.3637*** | 0.8655** | |
| Stock Market Index | -0.0079*** | ||
| Banking Institutions: | 0.5173*** | -0.0620 | |
| Deposits | |||
| Observations | 255 | 278 | 251 |
Bank and Market-Based Financing: Basic Specification
| (1) | (2) | (3) | |
|---|---|---|---|
| Market-based financing as share of GDP | Banking financing as share of GDP | Market-based as share of total financing | |
| GDP per capita in USD | 0.0022*** | 0.0024*** | 0.0000 |
| Domestic Savings | 0.8958*** | 0.8313** | |
| Public Bond Market | 0.2735*** | -0.5146*** | 0.4197*** |
| Capitalization | |||
| Capital Account | -10.9993** | -4.4179 | 0.4574 |
| Openness | |||
| T-Bond, 2 YRS | -1.3637*** | 0.8655** | |
| Stock Market Index | -0.0079*** | ||
| Banking Institutions: | 0.5173*** | -0.0620 | |
| Deposits | |||
| Observations | 255 | 278 | 251 |
The role of Crises
After financial crises, countries seem to experience a reduction in bank credit concomitantly with an increase in market-based financing. In order to test the effect of financial crises on the depth and structure of EMs credit markets, a step-dummy variable is introduced to the basic specification.14 The results suggest that after financial crises, EMs experience a reduction in bank credit and an increase in market-based financing. Financial crises, however, usually bring about significant changes in the functioning of banking systems and in regulatory frameworks, which cannot be easily disentangled.
A sequential analysis suggests that the dynamics of bank credit after financial crises is likely associated with a number of factors.15 The basic specification incorporates a crisis-dummy sequentially.16 The results suggest that the impact of crises is not necessarily related to the financial distress itself, but to changes that take place following the crises, (Figure 5 illustrates the results of the analysis for some EMs, including Mexico). Many factors may explain changes in banking sector credit after financial crises, including inter alia a lower level of savings in the form of bank deposits and the implementation of tighter prudential regulations.
Bank Credit to the Private Sector
(Share of GDP)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
Bank Credit to the Private Sector
(Share of GDP)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
Bank Credit to the Private Sector
(Share of GDP)
Citation: IMF Staff Country Reports 2012, 317; 10.5089/9781475543551.002.A004
Institutional Factors
An institutional variable is added to the specification. The World Bank’s World Governance Indicator (WGI) for rule of law is used, which captures perceptions of the extent to which agents have confidence in and abide the rules of society, and in particular the quality of contract enforcement, property rights, the police, and the courts, as well as the likelihood of crime and violence. The WB produces a score based on an unobserved components model for each economy, and then ranks them in a scale of 0 to 100.17
Better institutions have a positive effect on bank lending in EMs (Table 4). The new specification shows that the WGI variable is positive and significant in the regression of bank credit in EMs, which is likely related to an increase in the quality of contract enforcement and property rights. The crisis variable, however, becomes non-significant suggesting that the crisis step-dummy variable may be capturing an effect not directly related to financial distress during crises.
Bank and Market-Based Financing: Crises
Bank and Market-Based Financing: Crises
| (1) | (2) | (3) | |
|---|---|---|---|
| Market-based financing as share of GDP | Banking financing as share of GDP | Market-based as share of total financing | |
| GDP per capita in USD | 0.0017*** | 0.0025*** | -0.0002 |
| Domestic Savings | 0.7398** | 0.6774** | |
| Public Bond Market | 0.1993** | -0.4550*** | 0.3170*** |
| Capitalization | |||
| Capital Account | -7.3950* | -7.2752* | 4.0016* |
| Openness | |||
| T-Bond, 2 YRS | -0.9868*** | 0.6888** | |
| Stock Market Index | -0.0060*** | ||
| Crises | 7.0325*** | -6.6633** | 8.0622*** |
| Banking Institutions: | 0.6596*** | -0.2296*** | |
| Deposits | |||
| Observations | 255 | 278 | 251 |
Bank and Market-Based Financing: Crises
| (1) | (2) | (3) | |
|---|---|---|---|
| Market-based financing as share of GDP | Banking financing as share of GDP | Market-based as share of total financing | |
| GDP per capita in USD | 0.0017*** | 0.0025*** | -0.0002 |
| Domestic Savings | 0.7398** | 0.6774** | |
| Public Bond Market | 0.1993** | -0.4550*** | 0.3170*** |
| Capitalization | |||
| Capital Account | -7.3950* | -7.2752* | 4.0016* |
| Openness | |||
| T-Bond, 2 YRS | -0.9868*** | 0.6888** | |
| Stock Market Index | -0.0060*** | ||
| Crises | 7.0325*** | -6.6633** | 8.0622*** |
| Banking Institutions: | 0.6596*** | -0.2296*** | |
| Deposits | |||
| Observations | 255 | 278 | 251 |
Bank and Market-Based Financing: Institutional Factors
Bank and Market-Based Financing: Institutional Factors
| (1) | (2) | (3) | |
|---|---|---|---|
| Market-based financing as share of GDP | Banking financing as share of GDP | Market-based as share of total financing | |
| GDP per capita in USD | 0.0011*** | 0.0023*** | -0.0001 |
| Domestic Savings | 0.5115* | 0.1529 | |
| Public Bond Market | 0.3389*** | -0.6030*** | 0.3572*** |
| Capitalization | |||
| Capital Account | -0.3988 | -9.6663 | 5.1536* |
| Openness | |||
| T-Bond, 2 YRS | -0.9338*** | 0.3940 | |
| Stock Market Index | -0.0054*** | ||
| World Bank Governance | 0.1453 | 0.4950*** | -0.1143 |
| Indicator | |||
| Crises | 2.9188* | -4.0389 | 4.4026*** |
| Banking Institutions: | 0.5259*** | -0.2330*** | |
| Deposits | |||
| Observations | 222 | 239 | 220 |
Bank and Market-Based Financing: Institutional Factors
| (1) | (2) | (3) | |
|---|---|---|---|
| Market-based financing as share of GDP | Banking financing as share of GDP | Market-based as share of total financing | |
| GDP per capita in USD | 0.0011*** | 0.0023*** | -0.0001 |
| Domestic Savings | 0.5115* | 0.1529 | |
| Public Bond Market | 0.3389*** | -0.6030*** | 0.3572*** |
| Capitalization | |||
| Capital Account | -0.3988 | -9.6663 | 5.1536* |
| Openness | |||
| T-Bond, 2 YRS | -0.9338*** | 0.3940 | |
| Stock Market Index | -0.0054*** | ||
| World Bank Governance | 0.1453 | 0.4950*** | -0.1143 |
| Indicator | |||
| Crises | 2.9188* | -4.0389 | 4.4026*** |
| Banking Institutions: | 0.5259*** | -0.2330*** | |
| Deposits | |||
| Observations | 222 | 239 | 220 |
References
Armendariz de Aghion, B. and J. Morduch, 2005, “The Economics of Microfinance” Cambridge, MA: MIT Press.
Beck Thorsten, A. Demigurc-Kunt, and M. Martinez Peria, 2008, “Banking Services for Everyone? Barriers to Bank Access and Use around the World” The World Bank Economic Review, Volume 22 (3), 397–430.
Beck Thorsten, A. Demigurc-Kunt, and M. Martinez Peria, 2011, “Bank Financing for SMEs: Evidence Across Countries and Bank Ownership Types” Journal of Finance Services Research, 39, 35–54.
Berger, Allen, and Gregory Udell, 2006, “A More complete Conceptual Framework for SME Finance”, Journal of Banking and Finance, 30, 2945–2966.
Cull, Robert and Maria S. Martinez Peria, 2010, “Foreign Bank Participation in Developing Countries” World Bank Policy Research Working Paper 5398, World Bank.
De la Torre, A., A. Ize, and S. Schmukler, 2012, “Financial Development in Latin America and the Caribbean: The road Ahead” The World Bank.
De la Torre, A., M. Martinez Peria, and S. Schmukler, 2010, “Bank Involvement with SMEs: Beyond Relationship Lending” Journal of Banking and Finance, 34, 2280–2293.
Goldsmith, R., 1969, “Financial Structure and Development” New Heaven, CT: Yale University Press.
Levine, Ross, 1997, “Financial Development and Economic Growth: Views and Agenda” Journal of Economic Literature, Volume XXXV, 688–726.
Levine, Ross, 2005, “Finance and Growth: Theory and Evidence” Handbook of Economic Growth, Volume 1A.
Levine, Ross, A. Berger, A. Demirguc-Kunt, and J.G. Haubrich, 2004, “Bank Concentration and Competition: An Evolution in the Making” Journal of Money, Credit, and Banking, 36(3), 433–452.
Rajan, Raghuram and Luigi Zingales, 1998, “Financial Dependence and Growth” American Economic Review 88, 559–586.
Stultz, R., 2001, “Does Financial Structure Matters for Economic Growth? A Corporate Finance Perspective” In: Demirgüç-Kunt, A., Levine, R. (Eds.), Financial Structure and Economic Growth: A Cross-Country Comparison of Banks, Markets, and Development. MIT Press, Cambridge, MA, pp. 143–188.
Prepared by Francisco Vázquez Ahued and Esteban Vesperoni.
See, for example, Beck et.al. (2001), Goldsmith (1969), Levine (1997 and 2005), Rajan and Zingales (1998), Stultz (2001), and references therein.
For a summary on theoretical and empirical underpinnings in the relationship between financial development and growth, see Levine (2005).
These include relationship lending, credit scoring, standardized risk rating tools, asset-based lending, factoring, leasing, etc. See, for example, Berger and Udell (2006), and de la Torre, Martinez Pería and Schmukler (2010). Other papers on these issues are Armendariz et. al. (2005), Beck, Demigurc-Kunt and Martinez Pería (2008 and 2010); and Beck and Demigurc-Kunt (2007).
Assets under management by pension funds represented more than 10 percent of GDP by 2010.
These comparisons are based on a group of EMs in Latin America, Asia and Europe that have depicted dynamic financial systems during the last 20 years. The sample includes Brazil, Bulgaria, Chile, Colombia, Czech Republic, Estonia, Hungary, Indonesia, Korea, Latvia, Lithuania, Malaysia, Mexico, Peru, Philippines, Poland, Romania, Slovak Republic, Thailand, Turkey, and South Africa. In Figure 1, rest of the sample refers to non Latin American countries listed above.
This performance in banking system credit is consistent with the low level of intermediated funds. Bank deposits peaked during the 1990s at more than 30 percent of GDP, a level that was broadly similar to other EMs; and in particular to Latin American peers. However, the trend in deposits reversed in the mid-1990s, and bottomed out at about 20 percent of GDP in 2004. Deposits recovered to about 27 percent of GDP by 2010, but it is still well below the average level in other EMs (57 percent of GDP). Loan-to-deposit ratios in Mexico—at 1.2—are also lower than the average in other EMs at 1.65. Relatively low leverage may have also contained credit growth, although it has likely shielded the banking system from vulnerabilities experienced in other EMs.
The share of bank credit bottomed in 2004 at about 30 percent.
Although beyond the scope of this note, other factors at play could be related to competition in the banking sector.
For some countries, notably Mexico, pension system reforms after financial crises may have also played a role.
Some of the countries in the sample have not been included in the different regressions depending on the availability of information, in particular for bond and equity market instruments. The list of variables is described in Table 1.
The variable on equity financing has been adjusted for valuation changes, i.e. the variable captures the net financing associated to new issuances at a given year, valued in terms of the GDP of that year.
The following variables have been instrumented by using their lagged values: GDP per capita, domestic savings, public bond market capitalization, and capital account openness.
The variable takes the value zero as long as a country did not experience a financial crisis, and one afterwards.
The analysis was also performed for market-based financing instruments, with similar results.
Starting with a single 1-value dummy variable during the crisis year, we run multiple regressions incorporating an extra 1-value observation in subsequent years in each regression.
The World Bank produces this indicator bi-annually between 1994 and 1998. We interpolate the values of the indicator for 1995 and 1997.
