United States: Selected Issues

This Selected Issues paper focuses on U.S. potential growth in the aftermath of the crisis. It discusses recent productivity developments in the nonfarm business sector. The paper uses back-of-the envelope calculations to gauge the effect of diminished financial sector activity on GDP growth in coming years. A simple production function framework is used to give a long view of key factors explaining potential GDP growth in the United States in the last 30 years and explore possible developments in the next few years.

Abstract

This Selected Issues paper focuses on U.S. potential growth in the aftermath of the crisis. It discusses recent productivity developments in the nonfarm business sector. The paper uses back-of-the envelope calculations to gauge the effect of diminished financial sector activity on GDP growth in coming years. A simple production function framework is used to give a long view of key factors explaining potential GDP growth in the United States in the last 30 years and explore possible developments in the next few years.

II. Spillovers from U.S. Federal Debt Issuance: The Case of Emerging Market

Sovereign Borrowing

Oya Celasun

A. Introduction

1. Large projected federal budget deficits in the United States have led to concerns that large amounts of Treasury debt issuance may increase global interest rates and crowd out emerging market (EM) borrowing from global markets. The empirical literature suggests that an increase in publicly held U.S. federal debt of one percent of GDP raises long-term real U.S. Treasury debt yields by 3-4 basis points.1 A key question is how much of the increase in U.S. yields in response to higher U.S. federal public debt would be transmitted into yields on other instruments, including those issued by EM governments.

2. The yield on an EM sovereign bond can be decomposed into two parts; the yield on a corresponding U.S. Treasury bond and the spread between the yield on the EM sovereign bond and the corresponding U.S. Treasury bond. This chapter discusses the possible effects of U.S. Treasury debt issuance on the latter part—the spread between EM sovereign yields and U.S. Treasury debt yields.

B. Literature and Results

3. There are reasons to expect both negative and positive effects of U.S. federal debt on EM spreads. Empirical evidence points to a negative correlation between the amount of U.S. federal debt as a share of GDP and the spread between U.S. corporate and U.S. Treasury bond yields. Krishnamurthy and Vissing-Jorgensen (2008) argue that this correlation reflects a “convenience yield” on U.S. Treasury debt, the marginal valuation of which declines with the amount of outstanding debt. They attribute the convenience yield to the superior trading liquidity and low default risk of Treasury debt instruments, as well as their use in satisfying regulatory mandates. It is possible that this negative correlation also applies to the yield spread between EM sovereign bonds—often considered to be an asset class similar to U.S. high yield corporate bonds—and U.S. Treasury debt.

4. At the same time, as argued by Kamin and Kleist (1999), an increase in U.S. Treasury rates—an asset class with very little default risk—could be met by a bigger increase in yields on all risky assets, including EM debt, as investors seek compensation for the extra risk.2 Higher real benchmark rates could also raise perceptions of EM sovereign default risk given the larger debt servicing burden, thereby increasing spreads.

5. Regression analysis of EM sovereign spreads for the period 1997–2006 lends some support to the view that higher expected levels of U.S. federal debt (as a share of U.S. GDP) are associated with higher EM spreads. These types of regressions can face identification issues; in this context, changes in the level of federal public debt could capture a host of factors that can directly impact EM spreads—such as the business cycle and financial conditions in the United States. Accordingly, the regressions control for U.S. real GDP growth forecasts (from Consensus Economics), changes in real U.S. stock prices (as a forward looking indicator of U.S. activity), term premia on longer maturity U.S. Treasury debt, the VIX volatility index, and high-yield corporate spreads. Measures of EM sovereign creditworthiness, including credit ratings, external public debt as a share of GDP, real growth, equity price changes, and indices of political risk are included as controls.

6. The estimated effect of U.S. debt on EM spreads is statistically and economically significant (Table 1, column 1). The estimated equation implies that if projected U.S. federal debt were to increase by 20 percentage points of U.S. GDP, EM spreads would increase by about 8 percent. If the initial level of debt is 40 percent of GDP and all other explanatory variables are set to their sample mean, the resulting increase in spreads would be about 30 basis points. This effect would come on top of the roughly 60 basis points increase in Treasury yields that would be expected on the basis of a U.S. federal debt increase of that magnitude. The size of the estimated effect is robust to controlling for EM specific one-year ahead growth forecasts for the subset of the sample for which GDP forecasts are available (column 2). Higher yields on U.S. Treasury bonds, e.g. with five year maturity, are also found to be associated with higher EM spreads (column 3).3 This finding is consistent with the notion that the increase in EM yields on account of higher U.S. debt would be larger than the corresponding increase in U.S. Treasury yields.

Table 1.

The Determinants of Emerging Market Sovereign Spreads

article image
Source: Fund staff estimates.Notes: Robust standard errors clustered by country, ***, **, and * denote significance at 1, 5, and 10%, respectively. The dependent variable is the natural logarithm of EMBI Global sovereign stripped spreads (i.e. stripped of collateralized cash flows). The U.S. debt variable is a rolling weghted average of CBO baseline debt projections for the current and next years. All regressions include country fixed effects. Also included were the spread between five and one year Treasury yields, four and five year ahead U.S. real growth forecasts, and the U.S. high yield corporate bond spread; the coefficients for these variables are not shown as they were not significant in most specifications. EM sovereign spreads are from J.P. Morgan (EMBI Global bond index), all growth forecasts are from Consensus Economics. A higher value of the political risk index denotes lower risk.

7. Evidence on how prospective U.S. economic performance affects EM sovereign spreads is mixed. Near term indicators such as growth expectations for the current year and the changes in real stock prices over the past year possibly capture current global investor sentiment and are associated with lower spreads. By contrast, two-year ahead U.S. growth expectations or the term-premium on ten year Treasury bonds are weakly related to higher EM spreads, suggesting that demand for EM sovereign debt may be higher when expectations of medium-term U.S. growth are relatively weak.

C. Conclusions and Policy Implications

8. The regressions presented in this chapter suggest that a large increase in U.S. federal public debt is likely to put upward pressure on EM spreads. The effect of U.S. debt issuance could be moderated by stronger growth expectations in EMs relative to the United States, or actions that would lower EM sovereign risk, such as reducing external public debt. The findings reinforce the importance of implementing fiscal reforms and stabilizing federal public debt in the United States given its potential global spillover effects.

Table 2.

Estimated economic effects of selected explanatory variables

article image
Source: Staff calculations.Notes: For all variables except U.S. debt, calculations were done using the sample means as initial values. The means were about four percent for actual and expected real growth and 27 percent of GDP for external public debt in EMs. The initial value of the U.S. debt variable was set to 40 percent of GDP (its level at end-FY2008). The table presents the estimated effects of increases in U.S. debt of one and twenty percent of GDP, respectively, as shown in column 2. For all other variables, the calculations show the effect of increases of one percentage point versus increases equal to one sample standard deviation of the variable, as listed in the first column.

References

  • Engen, E., and G. Hubbard, 2004, “Federal Government Debt and Interest Rates,” NBER Macroeconomics Annual 2004, pp. 83138.

  • Kamin, S., and K. von Kleist, 1999, “The Evolution and Determinants of Emerging Market Credit Spreads in the 1990s,” Board of Governors of the Federal Reserve System, International Finance Discussion Papers, Number 653.

    • Search Google Scholar
    • Export Citation
  • Khrishnamurty, A., and A. Vissing-Jorgensen, 2008, “The Demand for Treasury Debt,” Manuscript, Northwestern University.

  • Laubach, T., 2009, “New Evidence on the Interest Rate Effects of Budget Deficits and Debt,” forthcoming in the Journal of the European Economic Association.

    • Search Google Scholar
    • Export Citation
1

Laubach (2009) identifies the relationship by estimating the effect of long-horizon forward rates (the five-year ahead 5- or 10-year forward rates) and future deficits projected by the Congressional Budget Office (under the assumption of unchanged laws and policies). He finds an effect of 3-4 basis points per one percentage point increase in the debt/GDP ratio. Engen and Hubbard (2004) test an array of specifications and conclude that the effect is about 3 basis points.

2

Consistent with this hypothesis, Kamin and Kleist (1999) find evidence of a positive relationship between three-month U.S. T-bill rates and EM Brady bond spreads.

3

The results are similar for yields on three- or ten-year Treasury bonds.