## 1. US Dollar interest rates: US 3-month Treasury yield, US 2-year Treasury

## Source details

**Canonical URL:** [1. US Dollar interest rates: US 3-month Treasury yield, US 2-year Treasury](https://www.imf.org/-/media/files/publications/wp/2018/wp1818.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2018/wp1818.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2018/wp1818.pdf.json)

---

### Data and methodology
- Data spans daily yields between January 2002 - December 2016 and covers both the pre- and post-crisis periods.
- Yield-to-maturity data from country bond indices (GBI-EM Broad and EMBI Global/Diversified) constructed and maintained by J.P. Morgan.
- Empirical exercise uses dollar-denominated yields of thirteen countries and local currency bond yields of fifteen countries.
  - Dollar-denominated group: Brazil, Chile, China, Colombia, Indonesia, Malaysia, Mexico, Peru, Philippines, Poland, Russia, South Africa, Turkey.
  - Local currency group: Brazil, Chile, China, Colombia, Hungary, India, Indonesia, Korea, Malaysia, Mexico, Poland, Russia, South Africa, Thailand, Turkey.
- Balanced samples required for factor analysis; countries with substantial missing data were dropped to prevent shortening of the sample period.
- The analysis correlates estimated common factors with five sets of international variables: US interest rates, Euro-area interest rates, regional and country equity indices, measures of liquidity and volatility, and commodity price indices.

### Results — EM sovereign dollar-denominated debt
- Factor structure:
  - Single significant factor (F1) drives the common variation in yields on sovereign dollar-denominated debt for the thirteen EM countries.
- Key quantitative findings:
  - F1 explains 79.5 percent (10.33 ÷ 13) of the total variability (sum of diagonal elements of the correlation matrix) in sovereign debt yields.
  - Average uniqueness is 0.2 (0.17 excluding China), implying about 80 percent (83 percent excluding China) of the total variability is accounted for by common factors.
  - Uniqueness ranges from 0.06 for Mexico to 0.62 for China:
    - Common factor accounts for 94 percent of variation in Mexico’s yields.
    - Common factor accounts for about 38 percent of variation in China’s yields.
- Regional decomposition (three sub-groups):
  - Sub-group 1 (Asia: China, Indonesia, Malaysia, Philippines): common factor accounts for 67 percent of total variability.
  - Sub-group 2 (Latin America: Brazil, Chile, Colombia, Mexico, Peru): common factor accounts for 86 percent of total variability.
  - Sub-group 3 (Poland, Russia, South Africa, Turkey): common factor accounts for 77 percent of total variability.
- Correlations with international variables (selected):
  - US Treasury correlation with F1: 0.81 (t-statistic (77.61)) for full sample; sub-groups: 0.84, 0.80, 0.74 (t-statistics: (86.50), (81.98), (66.98)).
  - US Treasury 3-months correlations: 0.59, 0.64, 0.41, 0.35 (t-statistics: (41.43), (47.02), (27.83), (22.84)).
  - US Treasury 2-years correlations: 0.69, 0.73, 0.55, 0.48 (t-statistics: (53.73), (60.74), (39.88), (33.58)).
  - US Treasury 10-years correlations: 0.81, 0.83, 0.77, 0.71 (t-statistics: (76.56), (83.86), (73.75), (61.75)).
  - MSCI Emerging Market Index correlations: -0.59, -0.57, -0.79, -0.75 (t-statistics: (-40.50), (-38.57), (-78.93), (-69.59)).
  - S&P GSCI Index correlations: -0.41, -0.37, -0.68, -0.63 (t-statistics: (-24.95), (-22.47), (-56.65), (-49.91)).
  - Libor-OIS correlation with F1 (full sample): 0.37 (t-statistic (22.12)).
- Interpretation:
  - US and Euro-area interest rates are important drivers of yields on EM sovereign dollar-denominated bonds.
  - US and Euro-area rates are more important in influencing Latin American dollar-denominated sovereign yields compared to the other sub-groups.

### Results — EM sovereign local currency debt
- Factor structure:
  - Three common factors (F1, F2, F3) affecting yields; common factors account for 11.068, which is 74 percent (11.068 ÷ 15) of the total variability.
- Proportion of common variance explained by factors (Table 6):
  - F1 = 0.627
  - F2 = 0.243
  - F3 = 0.129
- Country-level uniqueness and communalities (selected):
  - Korea: uniqueness 0.07; three common factors explain 93 percent of total variation.
  - South Africa: uniqueness 0.59; common factors explain 41 percent of total variation.
  - Countries with uniqueness < 0.2: Korea, Mexico, Chile, Poland, Indonesia, Colombia (common factors have strongest influence).
  - Countries with uniqueness between 0.2 and 0.3: India, Brazil, Turkey, Thailand, Hungary.
  - Countries with uniqueness > 0.3: South Africa, Russia, China, Malaysia.
- Factor descriptions and key loadings:
  - Factor 1 (international interest rate factor):
    - Accounts for about 63 percent of variability accounted for by common factors.
    - Nine countries with loadings greater than 0.7: Indonesia, Korea, Thailand, Chile, Colombia, Mexico, Hungary, Poland, Turkey.
    - Highly positively correlated with US and Euro-area yields.
    - Quantitative implication: international interest rates explain a larger proportion of total variability in EM sovereign dollar-denominated debt (80 percent) than in EM sovereign local currency debt (46 percent = 74 percent x 63 percent).
  - Factor 2 (commodities factor):
    - Accounts for 24 percent of variability explained by common factors.
    - Negatively correlated with the S&P GSCI commodity index, and the price indices for copper and crude-oil (Brent).
    - High positive loadings (greater than 0.5) from Brazil, Malaysia, Russia, South Africa; negative loading (less than -0.5) from Chile.
    - Interpretation: high commodity prices associated with low values for factor 2, implying relatively low interest rates in Brazil, Malaysia, Russia, South Africa.
  - Factor 3 (emerging China-India factor):
    - Explains 13 percent of variability due to common factors.
    - China and India are the only countries with loadings of about 0.7 on it.
    - Factor 3 is not highly correlated with the international variables considered; interpreted as an emerging China-India factor.
- Correlations with international variables (selected from Table 7):
  - US Treasury correlation with F1: 0.84 (t-statistic (85.05)).
  - US Treasury 10-years correlation with F1: 0.86 (t-statistic (92.26)).
  - Euribor correlations with F1: Euribor 3-months 0.87 (94.11); Euribor 2-years 0.91 (118.28); Euribor 10-years 0.93 (140.99).
  - Commodity correlations with F2: S&P GSCI Index F2 = -0.69 (t-statistic (-52.59)); WTI F2 = -0.64 (t-statistic (-45.76)); Copper F2 = -0.72 (t-statistic (-55.84)).
  - MSCI Emerging Market Index correlations with F1–F3: -0.29, -0.55, 0.51 (t-statistics: (-16.80), (-35.72), (32.18)).

### Evolution over time and crisis effects
- EM dollar-denominated debt (Table 8):
  - Pre-crisis (5/28/2004-9/12/2008): F1 = 52.2%; F2 = 36.7%; F3 = 11.1%
  - Post-crisis (9/15/2008-12/31/2016): F1 = 91.8%; F2 = 8.2%
  - Full sample: F1 = 100.0%
  - Common Variance: Pre-crisis = 11.791; Post-crisis = 11.033; Full = 10.332
  - Total Variance: 13
  - Observation: post-crisis period (2009-2016) F1 accounted for about 92 percent of the common variance.
- EM local currency debt (Table 9):
  - Pre-crisis (2/1/2005-9/12/2008): F1 = 42.0%; F2 = 28.7%; F3 = 21.4%; F4 = 7.8%
  - Post-crisis (9/15/2008-12/31/2016): F1 = 48.2%; F2 = 36.2%; F3 = 15.6%
  - Full sample: F1 = 62.7%; F2 = 24.3%; F3 = 12.9%
  - Common Variance: Pre-crisis = 12.644; Post-crisis = 12.245; Full = 11.068
  - Total Variance: 15
  - Observation: proportion of common variation explained by factor 1 rose from 33 percent in 2009 to 63 percent in 2016; factor 1 and factor 2 together accounted for 71 percent of common variation in the pre-crisis period and 84 percent in the post-crisis period.

### Policy implications and recommendations
- Main implications:
  - EM bond markets have become a global asset class with increased foreign investor participation, influenced strongly by global monetary conditions.
  - Increased integration raises liquidity but heightens sensitivity to external events; shocks from larger markets like the US and Eurozone may propagate more quickly and have wider global effects.
  - Issuance of sovereign and corporate debt in local currencies has benefits but may raise challenges related to EM debt levels, currency mismatches, and original sin.
- Measures to mitigate increased volatility and risks:
  - Broaden and deepen the domestic investor base (insurance companies, pension funds, investment funds) to enhance market liquidity and provide partial offset during capital outflows.
  - Incorporate stress scenarios of rollover risk from foreign investor holdings of domestic currency bonds into reserve adequacy measures.
  - Consider macro-prudential and capital flow management measures to address sovereign foreign exchange liquidity risk and influence inflow composition, including:
    - (i) higher reserve requirements, either on short-term external liabilities or on liabilities to non-resident investors;
    - (ii) minimum holding period for bonds, particularly for foreign investors in local currency bond markets;
    - (iii) taxes on foreign bond inflows.

*Source: wp1818 - 1. US Dollar interest rates: US 3-month Treasury yield, US 2-year Treasury (IMF working paper content provided).*

### 1. US Dollar interest rates: US 3-month Treasury yield, US 2-year Treasury

### 1. US Dollar interest rates: US 3-month Treasury yield, US 2-year Treasury

### Data and methodology
- Data spans daily yields between January 2002 - December 2016 and covers both the pre- and post-crisis periods.  
- Yield-to-maturity data from country bond indices (GBI-EM Broad and EMBI Global/Diversified) constructed and maintained by J.P. Morgan.  
- Empirical exercise uses dollar-denominated yields of thirteen countries and local currency bond yields of fifteen countries.  
  - Dollar-denominated group: Brazil, Chile, China, Colombia, Indonesia, Malaysia, Mexico, Peru, Philippines, Poland, Russia, South Africa, Turkey.  
  - Local currency group: Brazil, Chile, China, Colombia, Hungary, India, Indonesia, Korea, Malaysia, Mexico, Poland, Russia, South Africa, Thailand, Turkey.  
- Balanced samples required for factor analysis; countries with substantial missing data were dropped to prevent shortening of the sample period.  
- The analysis correlates estimated common factors with five sets of international variables: US interest rates, Euro-area interest rates, regional and country equity indices, measures of liquidity and volatility, and commodity price indices.

### Results — EM sovereign dollar-denominated debt
- Factor analysis indicates a single significant factor (F1) drives the common variation in yields on sovereign dollar-denominated debt for the thirteen EM countries.  
- Key quantitative findings:
  - F1 explains 79.5 percent (10.33 ÷ 13) of the total variability (sum of diagonal elements of the correlation matrix) in sovereign debt yields.  
  - Average uniqueness (part of total variation not explained by common factors) is 0.2 (0.17 excluding China), implying about 80 percent (83 percent excluding China) of the total variability is accounted for by common factors.  
  - Uniqueness ranges from 0.06 for Mexico to 0.62 for China:
    - Common factor accounts for 94 percent of variation in Mexico’s yields.
    - Common factor accounts for about 38 percent of variation in China’s yields.  
- Regional decomposition (three sub-groups):
  - Sub-group 1 (Asia: China, Indonesia, Malaysia, Philippines): common factor accounts for 67 percent of total variability.  
  - Sub-group 2 (Latin America: Brazil, Chile, Colombia, Mexico, Peru): common factor accounts for 86 percent of total variability.  
  - Sub-group 3 (Poland, Russia, South Africa, Turkey): common factor accounts for 77 percent of total variability.  
- Correlations with international variables:
  - Single common factor has a high positive correlation (defined as above 0.7) with US and Euro-area interest rates.  
  - Correlation with the US Treasury 10-year rate is high; correlation with the US Treasury 2-year rate is close to 0.7; correlation with the US Treasury 3-month rate is lower.  
  - Euro-area rates (3-month, 2-year, 10-year Euribor) also have significant correlations.  
  - Latin American sub-group exhibits strong correlation with copper and soybean prices.  
- Interpretation:
  - US and Euro-area interest rates are important drivers of yields on EM sovereign dollar-denominated bonds.
  - US and Euro-area rates are more important in influencing Latin American dollar-denominated sovereign yields compared to the other sub-groups.

### Results — EM sovereign local currency debt
- Factor analysis identifies three common factors (F1, F2, F3) affecting yields; common factors account for 11.068, which is 74 percent (11.068 ÷ 15) of the total variability.  
- Uniqueness measures vary considerably across countries:
  - Korea: uniqueness 0.07; three common factors explain 93 percent of total variation.  
  - South Africa: uniqueness 0.59; common factors explain 41 percent of total variation.  
  - Countries with uniqueness < 0.2: Korea, Mexico, Chile, Poland, Indonesia, Colombia (common factors have strongest influence).  
  - Countries with uniqueness between 0.2 and 0.3 (common factors less important): India, Brazil, Turkey, Thailand, Hungary.  
  - Countries with uniqueness > 0.3 (common factors least important): South Africa, Russia, China, Malaysia.  
- Factor descriptions and key loadings:
  - Factor 1 (international interest rate factor):
    - Accounts for about 63 percent of variability accounted for by common factors.
    - Nine countries with loadings greater than 0.7: Indonesia, Korea, Thailand, Chile, Colombia, Mexico, Hungary, Poland, Turkey.
    - Highly positively correlated with US and Euro-area yields, implying US and European monetary policies strongly influence local currency debt yields in these nine countries.
    - Quantitative implication: international interest rates explain a larger proportion of total variability in EM sovereign dollar-denominated debt (80 percent) than in EM sovereign local currency debt (46 percent = 74 percent x 63 percent).  
  - Factor 2 (commodities factor):
    - Accounts for 24 percent of variability explained by common factors.
    - Negatively correlated with the S&P GSCI commodity index, and the price indices for copper and crude-oil (Brent).
    - High positive loadings (greater than 0.5) from Brazil, Malaysia, Russia, South Africa; negative loading (less than -0.5) from Chile.
    - Interpretation: high commodity prices associated with low values for factor 2, implying relatively low interest rates in Brazil, Malaysia, Russia, South Africa.
  - Factor 3 (emerging China-India factor):
    - Explains 13 percent of variability due to common factors.
    - China and India are the only countries with loadings of about 0.7 on it.
    - Factor 3 is not highly correlated with the international variables considered; interpreted as an emerging China-India factor.

### Evolution over time and crisis effects
- Sequential factor analysis over 2004-2016:
  - EM dollar-denominated debt:
    - 2004-2008: three factors explained common variation.
    - By 2013: factor 1 (international interest rate factor) became dominant and the single factor accounting for common variation.
    - Post-crisis period (2009-2016): factor 1 accounted for about 92 percent of the common variance (Table 8).  
    - The global financial crisis did not impede the growing influence of international interest rates; dominance of factor 1 may partly reflect the crisis as a systemic shock.  
  - EM local currency debt:
    - 2005-2015: four or five factors explained common variation; by end-2016 only three factors were needed.
    - Proportion of common variation explained by factor 1 rose from 33 percent in 2009 to 63 percent in 2016.
    - Table 9: factor 1 (international interest rates) and factor 2 (commodity prices) together accounted for 71 percent of common variation in the pre-crisis period and 84 percent in the post-crisis period.

### Policy implications and recommendations
- Main implications:
  - EM bond markets have become a global asset class with increased foreign investor participation, influenced strongly by global monetary conditions.
  - Increased integration raises liquidity but heightens sensitivity to external events; shocks from larger markets like the US and Eurozone may propagate more quickly and have wider global effects.
  - Issuance of sovereign and corporate debt in local currencies has benefits but may raise challenges related to EM debt levels, currency mismatches, and original sin.  
- Measures to mitigate increased volatility and risks:
  - Broaden and deepen the domestic investor base (insurance companies, pension funds, investment funds) to enhance market liquidity and provide partial offset during capital outflows.
  - Incorporate stress scenarios of rollover risk from foreign investor holdings of domestic currency bonds into reserve adequacy measures.
  - Consider macro-prudential and capital flow management measures to address sovereign foreign exchange liquidity risk and influence inflow composition, including:
    - (i) higher reserve requirements, either on short-term external liabilities or on liabilities to non-resident investors;
    - (ii) minimum holding period for bonds, particularly for foreign investors in local currency bond markets;
    - (iii) taxes on foreign bond inflows.

*Source: wp1818 - 1. US Dollar interest rates: US 3-month Treasury yield, US 2-year Treasury (IMF working paper content provided).*

### References

### References

### Major thematic sources cited
- Literature on the global financial cycle, monetary policy spillovers, and implications for emerging markets: Arteta et al. (2015); Bernanke (2015); Borio (2014); Bruno and Shin (2015); Rey (2013, 2016); Miranda-Agrippino and Rey (2015); Rey (2013).
- Studies on emerging market (EM) bond markets, local-currency development, and foreign investor participation: Bae (2012); Ebeke and Lu (2015); Goswami and Sharma (2011); Jaramillo and Weber (2013a, 2013b); Miyajima, Mohanty, and Chan (2012); Peiris (2010); IMF (2016).
- Research on capital flows, bond yields, and spillovers from advanced-economy policies: Basu, Eichengreen, and Gupta (2014); Blanchard et al. (2016); Choi et al. (2017); Chari, Stedman, and Lundblad (2017); Fratzscher, Lo Duca, and Straub (2017); Singh and Wang (2017).
- Work on factor structure, common drivers and volatility in EM debt markets: McGuire and Schrijvers (2003, 2006); Bunda, Hamann, and Lal (2010); Fender, Hayo, and Neuenkirch (2012); Disyatat and Rungcharoenkitkul (2016); Drehmann, Borio, and Tsatsaronis (2012).

### Definitions and notes from the source
- Hard currency denotes funds that invest 75 percent or more in debt denominated in the following currencies: US Dollar, Euro, British Pound, Swiss Franc, Japanese Yen, Canadian Dollar, Australian Dollar, and Swedish Krona.

### Key figures and data points (as presented)
- Figure 3 — Allocations of U.S. investors to emerging markets (FY2015):
  - Emerging Market's Share of Global Output: 39.46
  - Allocation to Emerging Market Equity: 7.47
  - Allocation to Emerging Market Debt: 4.28
  - Allocation to Emerging Market Local Currency Debt: 1.39
- Figure 1 — Cumulative Net Inflow to EM Bonds (In USD Billions): charted series by Jan-04 to Jan-16 split into Hard Currency and Local Currency (no single totals provided in text).
- Figure 2 — Foreign Holdings of Local Currency Bonds (In percent of total LCY bonds): country series for JP, ID, KR, MY, TH across quarters 1998–2017 (plotted; no single numeric series printed in text).
- EM Local Currency and Dollar-denominated sovereign yields: time series plotted (Jan-02 to Jan-16) across listed countries (Brazil, Chile, China, Colombia, Hungary, India, Indonesia, Malaysia, Mexico, Peru, Philippines, Poland, Russia, South Africa, Thailand, Turkey, Korea) — graphs shown but no single numeric table of yields in the text.

### Factor analysis and variance decomposition — exact numeric results
- Table 1 (EM Sovereign Dollar-Denominated Debt, 2004-2016):
  - Variance accounted for by common factors: 10.332
  - Total Variance (sum of diagonal elements of correlation matrix): 13
- Table 2 (Sub-group 1, 2004-2016):
  - Variance accounted for by common factors: 2.677
  - Total Variance: 4
- Table 3 (Sub-group 2, 2002-2016):
  - Variance accounted for by common factors: 4.30
  - Total Variance: 5
- Table 4 (Sub-group 3, 2002-2016):
  - Variance accounted for by common factors: 3.062
  - Total Variance: 4
- Table 5 — Correlation between Common Factors and Other Variables (EM Sovereign Dollar-Denominated Debt):
  - US Treasury correlation with F1 (Full sample and sub-groups reported): 0.81, 0.84, 0.80, 0.74 (t-statistics shown in parentheses: (77.61), (86.50), (81.98), (66.98))
  - US Treasury 3-months correlations: 0.59, 0.64, 0.41, 0.35 (t-statistics: (41.43), (47.02), (27.83), (22.84))
  - US Treasury 2-years correlations: 0.69, 0.73, 0.55, 0.48 (t-statistics: (53.73), (60.74), (39.88), (33.58))
  - US Treasury 10-years correlations: 0.81, 0.83, 0.77, 0.71 (t-statistics: (76.56), (83.86), (73.75), (61.75))
  - Selected equity correlations (MSCI Emerging Market Index): -0.59, -0.57, -0.79, -0.75 (t-statistics: (-40.50), (-38.57), (-78.93), (-69.59))
  - Commodity correlations (S&P GSCI Index): -0.41, -0.37, -0.68, -0.63 (t-statistics: (-24.95), (-22.47), (-56.65), (-49.91))
  - Liquidity/volatility metrics (Libor-OIS, Ted Spread, VIX) reported with F1 correlations and t-statistics (e.g., Libor-OIS 0.37 with (22.12) for full sample grouping shown).
- Table 6 (EM Sovereign Local Currency Debt, 2005-2016):
  - Proportion of common variance explained by factors: F1 = 0.627, F2 = 0.243, F3 = 0.129
  - Variance accounted for by common factors: 11.068
  - Total Variance: 15
  - Country-level factor loadings and communalities shown for CHINA, INDIA, INDONESIA, KOREA, MALAYSIA, THAILAND, BRAZIL, CHILE, COLOMBIA, MEXICO, HUNGARY, POLAND, RUSSIA, TURKEY, SOUTH_AFRICA (exact loadings and communalities tabulated in the source).
- Table 7 — Correlation between Common Factors and Other Variables (EM Sovereign Local Currency Debt):
  - US Treasury correlation with F1: 0.84 (t-statistic (85.05)); US Treasury 10-years correlation with F1: 0.86 (t-statistic (92.26))
  - Euribor correlations with F1: Euribor 3-months 0.87 (94.11); Euribor 2-years 0.91 (118.28); Euribor 10-years 0.93 (140.99)
  - Equity correlations produce negative and mixed signs across F1–F3 (MSCI Emerging Market Index: -0.29, -0.55, 0.51 with t-statistics (-16.80), (-35.72), (32.18))
  - Commodity price correlations show strong negative correlations with F2 (e.g., S&P GSCI Index F2 = -0.69 with t-statistic (-52.59); WTI F2 = -0.64 with t-statistic (-45.76); Copper F2 = -0.72 with t-statistic (-55.84)).
- Table 8 (EM Sovereign Dollar-Denominated Debt, 2004-2016 — Proportion of Common Variance Explained by Factors):
  - Pre-crisis (5/28/2004-9/12/2008): F1 = 52.2%; F2 = 36.7%; F3 = 11.1%
  - Post-crisis (9/15/2008-12/31/2016): F1 = 91.8%; F2 = 8.2%
  - Full sample: F1 = 100.0%
  - Common Variance: Pre-crisis = 11.791; Post-crisis = 11.033; Full = 10.332
  - Total Variance: 13 (for all samples)
- Table 9 (EM Sovereign Local Currency Debt, 2005-2016 — Proportion of Common Variance Explained by Factors):
  - Pre-crisis (2/1/2005-9/12/2008): F1 = 42.0%; F2 = 28.7%; F3 = 21.4%; F4 = 7.8%
  - Post-crisis (9/15/2008-12/31/2016): F1 = 48.2%; F2 = 36.2%; F3 = 15.6%
  - Full sample: F1 = 62.7%; F2 = 24.3%; F3 = 12.9%
  - Common Variance: Pre-crisis = 12.644; Post-crisis = 12.245; Full = 11.068
  - Total Variance: 15 (for all samples)

### Observations implied by tables and figures (as presented in source)
- Common factors explain a large share of variance in EM dollar-denominated sovereign yields post-crisis: F1 accounts for 91.8 percent of common variance post-crisis (Table 8).
- For EM local-currency sovereign debt, the first factor (F1) explains a majority of common variance in the full sample (62.7 percent), with other factors (F2, F3) contributing materially (Table 9 and Table 6).
- Strong positive correlations exist between US Treasury yields and the primary common factor for both dollar-denominated and local-currency EM debt (e.g., US Treasury correlations with F1 commonly in the 0.8+ range in the presented tables).
- Commodity prices, equity indices, and volatility measures display substantial and heterogeneous correlations with the identified common factors across sub-groups and debt types (see Table 5 and Table 7 for exact correlations and t-statistics).

*Italic: Source — wp1818 - References (figures, tables, notes, and bibliography as provided in the source PDF).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2018/wp1818.pdf_