## _wp07228

## Source details

**Canonical URL:** [_wp07228](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07228.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07228.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07228.pdf.json)

---

### I. Introduction and research question
- Principal question: What are the main determinants of corporate bond spreads in emerging market economies (EMEs)?
- Objective: Disentangle the role of firm-specific variables, bond characteristics, observable macroeconomic conditions, sovereign risk, industry-specific effects and country-specific effects using a Bloomberg panel dataset with comparable corporate bond spreads.
- Motivation: Corporate bond yields reflect firms’ cost of financing; higher spreads raise the cost of capital and can reduce investment in physical capital, a key motor of economic growth.

### Data scope and sample
- Sample composition:
  - 139 bonds
  - 65 corporations
  - 10 EMEs
  - Regional breakdown: six in Latin America and four in the East Asian region
- Countries in sample: Argentina, Brazil, Chile, Colombia, Indonesia, Malaysia, Mexico, Philippines, Panama, Thailand.
- Panel: unbalanced over 29 consecutive quarters between 1999 and 2006.
- Currency: OAS of corporate bonds issued in foreign currencies only.
- Data sources:
  - Bloomberg OAS data for bond, firm, and global factor data.
  - World Bank’s World Development Indicators (WDI) for macro variables.
  - Sovereign risk proxied by EMBI plus (and EMBIglobal when EMBIplus unavailable).

### Methodology: OAS, model specification and estimation
- OAS rationale and computation:
  - OAS makes spreads from different bonds comparable by accounting for credit risk and contingent cash flow risk; defined as the spread added to the current short-term interest rate to make the theoretical bond price equal to the observed market price.
  - Bloomberg benchmark spot curve construction summarized in iterative steps (six-month spot rate benchmark, discount coupon cash flows, solve successive spot rates).
  - OAS for noncallable bonds: constant spread added to short-term rates using benchmark spot curve discount factors.
  - OAS for callable bonds: Bloomberg uses a one-factor arbitrage-free binomial tree of normally distributed short-rates driven by interest rate volatility; models call schedule and computes OAS across interest-rate scenarios.
- Empirical model (dependent variable: natural logarithm of OAS):
  - Specification components: country fixed effects, industry dummies, firm-level determinants, bond-structure characteristics, sovereign risk (CRct), country-level time-varying variables, global factors, error term.
  - Firm-level determinants (focus in reported regressions): EBIT/Assets, Equity/Assets (Equity / Capital), Debt/Assets, Size (Total Assets), Volatility Equity (360-day annualized standard deviation of daily log price changes).
  - Bond characteristics: Years to maturity (natural logarithm), interaction Years to Maturity x Debt/Assets to capture nonlinearity.
  - Country-level macro controls (preferred): GDP growth and GDP per capita.
  - Global factors: High Yield (US low-rated bond yield), Treasury 10 years (10-year U.S. treasury yield), VIX (CBOE implied volatility).
  - Sovereign risk: JPMorgan EMBI spread for each country.
- Econometrics:
  - Panel specifications exploit cross-section and time-series variation with country and industry fixed effects.
  - Standard errors clustered by “country x time” (Petersen, 2005).
  - Variance decomposition performed to quantify contribution of determinant sets.
  - Sensitivity tests: model in differences, additional interaction effects.

### Descriptive statistics (selected exact figures)
- Observations = 15292 for many variables.
- OAS spread: Mean 2.85, Std. Dev. 2.48.
- EBIT to assets: Mean 2.28, Std. Dev. 1.40.
- Equity to capital: Mean 48.58, Std. Dev. 18.47.
- Debt to assets: Mean 37.22, Std. Dev. 12.32.
- Assets: Mean 10,508, Std. Dev. 9,953.
- Volatility Equity: Mean 30.98, Std. Dev. 11.41.
- Years to Maturity: Mean 9.68, Std. Dev. 18.22.
- GDP Growth: Mean 4.01, Std. Dev. 2.62.
- GDP per capita: Mean 4,733, Std. Dev. 1,408.
- EMBI: Mean 2.65, Std. Dev. 4.53.
- High Yield: Mean 9.50, Std. Dev. 2.06.
- Treasury 10 years: Mean 4.54, Std. Dev. 0.62.
- VIX: Mean 17.55, Std. Dev. 6.76.

### Main empirical findings — determinants and magnitudes
- Determinants identified: firm-specific variables, bond characteristics, macroeconomic conditions, sovereign risk, global factors, industry and country fixed effects.
- Firm-level characteristics account for the largest share of variance in corporate OAS.
- Sovereign–corporate link:
  - Positive and significant “transfer risk” from sovereign spreads to corporate spreads, but less than one to one: a 1 percent increase in the default risk of the sovereign translates into a less than 1 percent increase in the default risk of the corporate sector.
  - Wald test for EMBI=1: EMBI 0.382***, Std. Err. 0.074, Wald Chi-square 68.2, Prob. 0.000 — rejects strict sovereign ceiling hypothesis.
- Asymmetry in spread responses: corporate spreads respond more acutely to increases in risk, consistent with investor herding and panic behavior in emerging markets.

### Representative coefficient estimates (examples from full specification / Column (9) and other referenced columns)
- Firm-level coefficients (Column (9) context):
  - EBIT/Assets: -0.043*** (Std. Err. 0.012)
  - Equity / Capital: -0.009*** (0.003)
  - Debt/Assets: 0.014*** (0.004)
  - Size: -0.180*** (0.022)
  - Volatility Equity: 0.016*** (0.002)
- Bond characteristics:
  - Years to Maturity: 0.473*** (0.059)
  - Years to Maturity x Debt/Assets: -0.005*** (0.001)
- Sovereign risk:
  - EMBI (Column (4) baseline): 0.382*** (0.075)
  - EMBI in full specification (Column (9)): 0.308*** (0.110)
  - EMBI x Years to Maturity: often negative and significant (example reported: -0.084*** (0.022) in some specifications).
- Macroeconomic controls:
  - GDP Growth: -0.014** (0.006) in full specification.
  - GDP per capita: reported but not statistically significant in preferred specification (example: 0.317 (0.434)).
- Global factors:
  - High Yield: 0.386*** (0.080) in Column (9).
  - Treasury 10 years: -0.418*** (0.100) when included.
  - VIX: 0.006* (0.004) marginally significant in some specifications.
- Model fit (R-squared progression Columns (1) to (9)): 0.474, 0.585, 0.671, 0.716, 0.718, 0.722, 0.743, 0.749, 0.75.

### Variance decomposition (based on column (9) without interaction terms)
- Relative contributions to total variance of corporate OAS (Figure 1 summary):
  - Firm-specific factors: approximately 33 percent of total variance.
    - Within firm-specific: Volatility of equity and Size account for approximately 20 percent altogether.
  - Fixed effects (industry-level and country-level): approximately 12 percent.
  - Global factors: approximately 9 percent.
  - Sovereign risk: approximately 9 percent.
  - Bond characteristics: 6 percent.
  - Macro-variables: 4 percent.
- Interpretation: firm-level characteristics are the dominant drivers of corporate spreads in EMEs; sovereign risk and global factors are quantitatively less important but still relevant.

### Extensions, asymmetries and robustness checks
- Sector interactions with EMBI (Table 2 highlights):
  - EMBI when interacting with tradable-sector dummy: EMBI = 0.196** (0.090) in Column (1).
  - EMBI x 1(Tradable): -0.024 (0.038) (not statistically significant in full specification).
  - Sector-specific mitigations of EMBI effect reported:
    - Tourism: EMBI x 1(Tourism) = -0.285*** (0.079)
    - Telecommunication: EMBI x 1(Telecommunication) = -0.473*** (0.117)
    - Construction: EMBI x 1(Construction) = -0.290* (0.164) (weak significance)
- First-differences model (Table 3) — asymmetries:
  - dEMBI: example 0.167** (0.071) in Column (1).
  - dHigh Yield: example 0.457*** (0.097).
  - Evidence of asymmetry: dEMBI x 1(dEMBI>0) = 0.046*** (0.017) in Column (2); dHigh Yield x 1(dHigh Yield>0) = 0.049** (0.020) in Column (3).
  - Interpretation: larger impact when EMBI or High Yield increases versus decreases, consistent with a “sovereign ceiling lite” and with panic/herding behavior.
- Additional robustness checks:
  - Excluding Argentina (Table 4): EMBI effect increases from 0.382 to 0.452.
  - Replacing EMBI with sovereign credit ratings (Table 5): qualitatively similar results; ratings less informative than spreads.
  - Distributional check: allowing EMBI elasticity to vary by bond finds that in 73 percent of cases elasticities lie between 0 and 1; mean 0.39 and median 0.36.

### Conclusions and policy-relevant implications
- Determinants of corporate OAS in EMEs include: firm-specific variables, country and industry fixed effects, sovereign risk, bond characteristics, macroeconomic conditions, and global factors.
- Firm-specific characteristics account for the largest share of variance in corporate spreads; management and firm fundamentals are critical determinants of corporate borrowing costs.
- Sovereign risk has a positive spillover effect on corporate spreads but not one-to-one; evidence rejects a strict sovereign ceiling while supporting a “sovereign ceiling lite.”
- Asymmetries: increases in sovereign risk and in U.S. High Yield have larger effects than decreases, signaling potential investor herding/panic dynamics in EMEs.
- Robustness: principal results hold under exclusion of Argentina, alternative sovereign risk measures (ratings), first-differences specification, and sectoral interaction tests.

*Source: _wp07228 - Section IV we present our regression results, along with a variance decomposition analysis to (IMF working paper content provided).*

### References..............................................................................................................

### _wp07228 - References..............................................................................................................

### I. Introduction and research question
- Principal question: What are the main determinants of corporate bond spreads in emerging market economies (EMEs)?
- Objective: Disentangle the role of firm-specific variables, bond characteristics, observable macroeconomic conditions, sovereign risk, industry-specific effects and country-specific effects using a Bloomberg panel dataset with comparable corporate bond spreads.
- Motivation: Corporate bond yields reflect firms’ cost of financing; higher spreads raise the cost of capital and can reduce investment in physical capital, a key motor of economic growth.

### Data scope and sample
- Sample size and composition:
  - 139 bonds
  - 65 corporations
  - 10 EMEs
  - Regional breakdown: six in Latin America and four in the East Asian region
- Data source: Bloomberg OAS data; sovereign risk proxied by EMBI plus (and EMBIglobal when EMBIplus unavailable).

### Methodology: OAS and comparison to sovereign spreads
- Option-Adjusted Spread (OAS) rationale:
  - OAS makes spreads from different bonds comparable by accounting for credit risk and contingent cash flow risk.
  - OAS is defined as the spread that must be added to the current short-term interest rate to make the “theoretical” price of the corporate bond identical to the observed market price (Choudhry, 2004).
  - Two reasons for using OAS instead of standard yield-to-maturity spreads:
    - Bonds with embedded options have cash flows dependent upon future interest rates, making definite maturity dates impractical.
    - Yield-to-maturity does not decompose how much of a bond’s spread is due to credit risk versus option risk; OAS allows this assessment against the risk-free “term structure of interest rates” or “benchmark spot curve.”
- Sovereign risk measure:
  - EMBI plus is used as the benchmark measure of sovereign risk in EMEs; includes U.S. dollar-denominated Brady bonds, Eurobonds, and traded loans issued by sovereign entities.
  - JPMorgan computes country spreads as a weighted average of spreads from bonds meeting size and liquidity requirements.
  - For the sample, EMBIplus is used mostly; EMBIglobal used when EMBIplus is unavailable (differences checked and not statistically significant where both exist).

### Main empirical findings
- Determinants identified:
  - Firm-specific variables
  - Bond characteristics
  - Macroeconomic conditions
  - Sovereign risk
  - Global factors
- Variance decomposition result:
  - Firm-level characteristics account for the largest share of the variance in corporate OAS.
- Sovereign–corporate link:
  - Evidence of a positive and significant “transfer risk” from sovereign spreads to corporate spreads, but less than one to one (described as a sovereign ceiling “lite”).
  - Quantitative statement preserved: a 1 percent increase in the default risk of the sovereign translates into a less than 1 percent increase in the default risk of the corporate sector.
- Asymmetry in spread responses:
  - Corporate spreads respond more acutely to increases in risk, consistent with investor herding and panic behavior in emerging markets.

### Contributions relative to prior literature
- Methodological contributions:
  - Use of OAS to study determinants of corporate bond spreads in a cross-country, multi-bond EME setting.
  - Application of panel data techniques to disentangle firm-specific, bond-specific, macroeconomic, sovereign, industry, and country effects.
- Relation to selected prior studies:
  - Differs from Huang and Kong (2003) by focusing on EMEs and multiple bonds.
  - Differs from Durbin and Ng (2005) by using OAS and examining a broader set of determinants beyond sovereign risk.
  - Differs from Peter and Grandes (2005) by sample scope (not limited to publicly traded South African firms in local currency) and by use of OAS.
  - Complements Borensztein et al. (2006a, 2006b) by addressing sovereign ceiling implications in terms of bond spreads (rather than ratings).

### Structure of the paper (as provided)
- Next sections (as indicated): empirical strategy and choice of variables; data and sample issues.

*Source: _wp07228 - References..............................................................................................................*

### Section IV we present our regression results, along with a variance decomposition analysis to

### _wp07228 - Section IV we present our regression results, along with a variance decomposition analysis to

### Empirical strategy
- Model: natural logarithm of the spread (Sit) specified as:
  - 01 2 3 45 6itCsititctcttit SPDFBCRCGβββββ β βμ=++++ +++
  - Components: country fixed effects (PC), industry dummies (Ds), firm-level determinants (Fit), bond-structure characteristics (Bit), sovereign risk (CRct), country-level time-varying variables (Cct), global factors (Gt), error μit.
- Dependent variable: OAS (in natural logarithm).
- Firm-level determinants (based on Altman (2000), with focus in reported regressions on):
  - EBIT/Assets, Equity/Assets, Debt/Assets, Size (Total Assets).
  - Also include Volatility Equity (standard deviation of day-to-day logarithmic price changes; 360-day annualized defined in note).
- Bond characteristics: Years to maturity (natural logarithm) and interaction Years to Maturity x Debt/Assets to capture nonlinearity.
- Country-level macro controls (preferred specification): GDP growth and GDP per capita.
- Global factors: High Yield (US low-rated bond yield), Treasury 10 years (10-year U.S. treasury yield), VIX (CBOE implied volatility).
- Sovereign risk: JPMorgan EMBI spread for each country.
- Econometrics:
  - Panel specifications exploit cross-section and time-series variation.
  - Country and industry fixed effects included (time-dummies not included in preferred specification; results robust to their inclusion).
  - Standard errors clustered by “country x time” (Petersen, 2005).
  - Variance decomposition performed to quantify contribution of determinant sets.
  - Sensitivity tests: model in differences, additional interaction effects.

### Data description
- Sample:
  - Panel of 139 bonds issued by 65 corporations across 10 EMEs between 1999 and 2006.
  - Countries: Argentina, Brazil, Chile, Colombia, Indonesia, Malaysia, Mexico, Philippines, Panama, Thailand.
  - Unbalanced panel over 29 consecutive quarters.
- Currency: OAS of corporate bonds issued in foreign currencies only.
- Data sources:
  - Bloomberg for bond, firm, and global factor data.
  - World Bank’s World Development Indicators (WDI) for macro variables.
- Descriptive statistics (selected, as reported in Table 8):
  - Observations = 15292 for many variables.
  - OAS spread: Mean 2.85, Std. Dev. 2.48.
  - EBIT to assets: Mean 2.28, Std. Dev. 1.40.
  - Equity to capital: Mean 48.58, Std. Dev. 18.47.
  - Debt to assets: Mean 37.22, Std. Dev. 12.32.
  - Assets: Mean 10,508, Std. Dev. 9,953.
  - Volatility Equity: Mean 30.98, Std. Dev. 11.41.
  - Years to Maturity: Mean 9.68, Std. Dev. 18.22.
  - GDP Growth: Mean 4.01, Std. Dev. 2.62.
  - GDP per capita: Mean 4,733, Std. Dev. 1,408.
  - EMBI: Mean 2.65, Std. Dev. 4.53.
  - High Yield: Mean 9.50, Std. Dev. 2.06.
  - Treasury 10 years: Mean 4.54, Std. Dev. 0.62.
  - VIX: Mean 17.55, Std. Dev. 6.76.

### OAS computation (Bloomberg Professional Analysis)
- Benchmark spot curve construction steps summarized:
  1. Six-month spot rate set to benchmark six-month rate.
  2. One-year benchmark bond with a six-month coupon considered.
  3. Discount six-month coupon using six-month spot rate to obtain present value.
  4. Subtract present value from market price to solve one-year spot rate.
  5. Repeat to obtain spot rates for successive terms.
- OAS for noncallable bonds: constant spread added to short-term rates to equate model price to market price using benchmark spot curve discount factors.
- OAS for callable bonds: Bloomberg uses a one-factor arbitrage-free binomial tree of normally distributed short-rates driven by interest rate volatility; models call schedule and computes OAS across interest-rate scenarios.

### Main regression results (Table 1 highlights)
- General: Sequential estimation of equation (1) with incremental inclusion of controls. Country and sector dummies included throughout.
- Key firm-level coefficients (examples from Column (9) full specification):
  - EBIT/Assets: -0.043*** (Std. Err. 0.012)
  - Equity / Capital: -0.009*** (0.003)
  - Debt/Assets: 0.014*** (0.004)
  - Size: -0.180*** (0.022)
  - Volatility Equity: 0.016*** (0.002)
- Bond characteristics:
  - Years to Maturity: 0.473*** (0.059)
  - Years to Maturity x Debt/Assets: -0.005*** (0.001)
- Sovereign risk:
  - EMBI (Column (4) baseline): 0.382*** (0.075)
  - EMBI in full specification (with global factors, Column (9)): 0.308*** (0.110) [noting drop in size when global factors included]
  - EMBI x Years to Maturity: negative and significant in many specifications (e.g., -0.084*** (0.022) in some columns), indicating larger EMBI impact on shorter maturities.
- Macroeconomic controls:
  - GDP Growth: -0.014** (0.006) in full specification (Column (9) context).
  - GDP per capita: coefficients reported but not statistically significant in preferred specification (e.g., 0.317 (0.434) in some columns).
- Global factors:
  - High Yield: positive and significant (e.g., 0.386*** (0.080) in Column (9)).
  - Treasury 10 years: negative and significant when included (e.g., -0.418*** (0.100) in column including treasury).
  - VIX: 0.006* (0.004) marginally significant in some specifications.
- Model fit:
  - R-squared progression across columns: 0.474, 0.585, 0.671, 0.716, 0.718, 0.722, 0.743, 0.749, 0.75 (Columns (1) to (9)).

### Variance decomposition (based on column (9) without interaction terms)
- Relative contributions to total variance of corporate OAS (Figure 1 summary):
  - Firm-specific factors: approximately 33 percent of total variance.
    - Within firm-specific: Volatility of equity and Size account for approximately 20 percent altogether.
  - Fixed effects (industry-level and country-level): approximately 12 percent.
  - Global factors: approximately 9 percent.
  - Sovereign risk: approximately 9 percent.
  - Bond characteristics: 6 percent.
  - Macro-variables: 4 percent.
- Interpretation: firm-level characteristics are the dominant drivers of corporate spreads in EMEs; sovereign risk and global factors are quantitatively less important but still relevant.

### Extensions and robustness checks
- Sector interactions with EMBI (Table 2):
  - EMBI coefficient when interacting with tradable-sector dummy: EMBI = 0.196** (0.090) in Column (1).
  - EMBI x 1(Tradable): -0.024 (0.038) (not statistically significant in full specification).
  - Sector-specific significant mitigations of EMBI effect observed for:
    - Tourism: EMBI x 1(Tourism) = -0.285*** (0.079)
    - Telecommunication: EMBI x 1(Telecommunication) = -0.473*** (0.117)
    - Construction: EMBI x 1(Construction) = -0.290* (0.164) (weak significance)
- First-differences model (Table 3) — asymmetries:
  - Changes in EMBI (dEMBI): positive and significant (examples: 0.167** (0.071) in Column (1)).
  - Changes in High Yield (dHigh Yield): positive and significant (examples: 0.457*** (0.097)).
  - Evidence of asymmetry: interaction dEMBI x 1(dEMBI>0) = 0.046*** (0.017) in Column (2) indicates larger impact when EMBI increases versus decreases.
  - Similar asymmetry for High Yield: dHigh Yield x 1(dHigh Yield>0) = 0.049** (0.020) in Column (3).
  - Interpretation: asymmetric responses consistent with a “sovereign ceiling lite” and with panic/herding behavior in EMEs.
- Additional robustness checks:
  - Excluding Argentina (Table 4): EMBI effect increases from 0.382 to 0.452 in robustness regressions.
  - Replacing EMBI with sovereign credit ratings (Table 5): qualitatively similar results; ratings less informative than spreads.
  - Distributional check: allowing EMBI elasticity to vary by bond finds that in 73 percent of cases elasticities lie between 0 and 1; mean 0.39 and median 0.36, close to benchmark estimate.

### Key quantitative findings (selected exact figures)
- Firm-level largest contributors to variance: approximately 33 percent total variance.
- Volatility of equity and Size jointly: approximately 20 percent of variance.
- Fixed effects: approximately 12 percent.
- Global factors: approximately 9 percent.
- Sovereign risk: approximately 9 percent.
- Bond characteristics: 6 percent.
- Macro-variables: 4 percent.
- EMBI point estimates:
  - Baseline: 0.382*** (Std. Err. 0.075) (Column (4) baseline).
  - Wald test for EMBI=1: EMBI 0.382***, Std. Err. 0.074, Wald Chi-square 68.2, Prob. 0.000 (Table 9) — rejects strict sovereign ceiling hypothesis.
- Representative coefficient magnitudes from full specification (Column (9) context):
  - EBIT/Assets: -0.043*** (0.012)
  - Equity / Capital: -0.009*** (0.003)
  - Debt/Assets: 0.014*** (0.004)
  - Size: -0.180*** (0.022)
  - Volatility Equity: 0.016*** (0.002)
  - Years to Maturity: 0.473*** (0.059)
  - Years to Maturity x Debt/Assets: -0.005*** (0.001)
  - High Yield: 0.386*** (0.080)
  - Treasury 10 years: -0.418*** (0.100)
  - VIX: 0.006* (0.004)
- Correlations (Table 6):
  - OAS–EMBI: 0.46
  - OAS–High Yield: 0.42
  - OAS–VIX: 0.43
  - High Yield–VIX: 0.87
  - Treasury 10y correlations low with others (e.g., OAS–Treasury 10y: 0.06)

### Conclusions
- Determinants of corporate OAS in EMEs include: firm-specific variables, country and industry fixed effects, sovereign risk, bond characteristics, macroeconomic conditions, and global factors.
- Firm-specific characteristics account for the largest share of variance in corporate spreads; management and firm fundamentals are critical.
- Sovereign risk has a positive spillover effect on corporate spreads but not one-to-one (evidence against a strict sovereign ceiling; consistent with a “lite” sovereign ceiling).
- Asymmetries: increases in sovereign risk and in U.S. High Yield have larger effects than decreases, consistent with panic/herding behavior in EMEs.
- Robustness: results hold under exclusion of Argentina, alternative sovereign risk measures (ratings), first-differences specification, and sectoral interaction tests.

*Source: _wp07228 - Section IV we present our regression results, along with a variance decomposition analysis to (IMF working paper content provided).*

### References

### _wp07228 - References

### Academic articles and working papers
- Altman, E., 2000, “Predicting Financial Distress of Companies: Revisiting the Z-score and Zeta® Models,” Working Paper, (New York: New York University, Stern School of Business), http://pages.stern.nyu.edu/~ealtman/Zscores.pdf
- Bandeira, O., et al., 2000, “Does Financial Reform Raise or Reduce Savings?” Review of Economics and Statistics, No. 82, pp. 239-63.
- Borensztein, E., K. Cowan, and P. Valenzuela, 2006a, “Sovereign Ceiling ‘Lite’? The Impact of Sovereign Ratings on Corporate Ratings in Emerging Market Economies,” IMF Working Paper 07/75 (Washington: International Monetary Fund).
- ———, 2006b, “The ‘Sovereign Ceiling Lite’ and Bank Credit Ratings in Emerging Markets Economies,” Mimeo, (Washington: Inter-American Development Bank). http://sechi.facea.uchile.cl/sechi/contributed_2/Valenzuela_patr.pdf
- Calvo, G., 1999, “Contagion in Emerging Markets: When Wall Street is a Carrier,” (Maryland: University of Maryland), http://www.bsos.umd.edu/econ/ciecrp8.pdf
- Calvo, G., and E. Mendoza, 2000, “Capital Markets Crises and Economic Collapse in Emerging Markets: An Informational Frictions Approach,” American Economic Review, No. 90 Vol. 2, pp. 59-94.
- Campbell, J., and G. Taksler, 2003, “Equity Volatility and Corporate Bond Yields,” Journal of Finance, No. 58 Vol. 6, pp. 2321-50.
- Ederington, H.L., J.B. Yawitz, and B.E. Robert, “The Information Content of Bond Ratings,” Journal of Financial Research, No. 10, pp. 211-26.
- Eichengreen, B., R. Hausmann, and U. Panizza, 2003, “Currency Mismatches, Debt Intolerance and Original Sin: Why They Are Not the Same and Why it Matters,” NBER Working Papers 10036, (Cambridge, Massachusetts: National Bureau of Economic Research).
- Durbin, E., and D. Ng, 2005, “The Sovereign Ceiling and Emerging Market Corporate Bond Spreads,” Journal of International Money and Finance No. 24 pp. 631-49.
- Huang, J., and W. Kong, 2003, “Explaining Credit Spread Changes: New Evidence from Option-Adjusted Bond Indexes,” Stern School of Business Finance Paper No. 03-013, (New York: New York University).
- Merton, R., 1974, “On the Pricing of Corporate Debt: The Risk Structure of Interest Rates,” Journal of Finance No. 29 Vol. 2, pp. 449-70.
- Peter, M., and M. Grandes, 2005, “How Important Is Sovereign Risk in Determining Corporate Default Premia? The Case of South Africa,” IMF Working Paper 05/217, (Washington: International Monetary Fund).
- Petersen, M., 2005, “Estimating Standard Errors in Finance Panel Data Sets: Comparing Approaches,” NBER Working Paper 11280, (Cambridge, Massachusetts: National Bureau of Economic Research).
- Standard and Poor’s, 2001, “Rating Methodology: Evaluating the Issuer, Corporate Ratings Criteria,” (New York: Standard and Poor’s).

### Books and institutional reports
- Choudhry, M. 2004, Advanced Fixed Income Analysis, (Amsterdam, The Netherlands: Elsevier Butterworth-Heinemann).
- IDB, 2006, Living with Debt: How to Limit the Risk of Sovereign Finance, Economic and Social Progress in Latin America: 2007 Report, (Washington: Inter-American Development Bank).

*Source: _wp07228 - References*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07228.pdf_