This Selected Issues paper reviews the benefits and costs of reserves for Chile, with emphasis on standard methodologies for assessing reserve adequacy. It reports an empirical methodology that analyzes simultaneously key explanatory variables behind a country’s level of reserves. The paper previews mechanisms that could supplement a country’s liquidity needs in times of stress. This paper also describes the instruments available for hedging foreign exchange risk, the demand and supply of foreign exchange hedging in the onshore market, and assesses foreign exchange exposure in different industrial sectors using factor analysis.

Abstract

This Selected Issues paper reviews the benefits and costs of reserves for Chile, with emphasis on standard methodologies for assessing reserve adequacy. It reports an empirical methodology that analyzes simultaneously key explanatory variables behind a country’s level of reserves. The paper previews mechanisms that could supplement a country’s liquidity needs in times of stress. This paper also describes the instruments available for hedging foreign exchange risk, the demand and supply of foreign exchange hedging in the onshore market, and assesses foreign exchange exposure in different industrial sectors using factor analysis.

III. Chile’s Experience with Inflation-Linked Bonds1

1. The Chilean experience with indexation is often associated with the successful development of long-term fixed-income markets. Several studies have pointed to the various benefits of the indexation of debt in terms of market completion, in particular in the context of the transition from a high-inflation to a low-inflation environment.2 In fact, Chile’s CPI inflation-based measure of account, Unidad de Fomento (UF), has constituted a central piece of the technology developed for protecting capital market participants from high inflation, providing investors with a unique set of bond return patterns and allowing borrowers to extend the duration of their liabilities.

2. In this regard, the monetary authorities in Chile have played a key role in the development of the inflation-indexed debt market. Until recently, most financial market and monetary policy instruments in Chile were denominated in UF. The central bank had extensively issued debt in UF, creating pricing benchmarks at various maturities facilitating the issuance of domestic corporate bonds in UF. However, with the shift to a nominal monetary policy target in 2001, the central bank has gradually moved to reduce the prevalence of indexation in financial markets and promote the development of a peso bond market.

3. With the shift to a nominal monetary policy target, important questions have been raised regarding the structure of the liabilities of the central bank and its implications for the local capital market. Given competing efforts to promote a liquid market of peso-denominated central bank paper, a natural policy issue pertains to the value of maintaining a well-functioning market for UF-denominated securities. In addition, in the current low inflation environment and with inflation expectations well-anchored around the middle of the inflation target bank, the role of the UF has become less evident. While a broader set of financial markets has obvious benefits, the development of these markets is costly and involves trade-offs related to economies of scale in liquidity and transaction costs.3

4. This chapter reviews the role of the central bank’s debt program in the context of the current low inflation environment. The analysis provides, first, a review of the recent development of the global market for inflation-linked securities in low-inflation economies and compares the degree of integration of the Chilean UF market with international markets. The analysis then points to one of the ancillary merits of maintaining a well-functioning UF bond market, in particular with regard to the real yield curve’s ability to predict future economic activity. Finally, the study reviews the implications of the shift toward a nominal monetary policy target and the changing structure of central bank debt for the development of the corporate bond market and integration with international markets.

A. The Global Market of Inflation-Linked Bonds and the Chilean Market

5. In the past decade, sovereign issuances of inflation-linked (IL) bonds have experienced a remarkable growth. Following the lead of the United Kingdom (1981) and Australia (1985), new markets were established in Canada (1991), Sweden (1994), the United States (1997), France (1998), South Africa (2000), and most recently in Greece (2003), Italy (2003), and Japan (2004). Over the past three years, the issuance of IL bonds quadrupled bringing the total global market capitalization of sovereign IL bonds close to US$ 500 billion by end 2003, nearly five times its size in 1996. The United States and the United Kingdom accounted for 68 percent of the market by end 2003, with a large presence in the medium- and long- range of the market (Figure 1). Because many major countries have relatively immature markets and will continue to expand their market for IL instruments toward a higher target, IL bonds are expected to grow at a rapid pace in coming years.4

Figure 1a.
Figure 1a.

Sovereign Inflation-Linked Bond Markets

Citation: IMF Staff Country Reports 2004, 292; 10.5089/9781451807608.002.A003

Source: Barclay’s Capital.
Figure 1b.
Figure 1b.

Yield Curves of Sovereign Inflation-Linked Bond Markets

Citation: IMF Staff Country Reports 2004, 292; 10.5089/9781451807608.002.A003

6. The increase in global supply and number of sovereign issuers has caused IL bonds to evolve into a global asset class. The rapid expansion of these instruments has met a growing appetite by investors seeking to enhance scope for diversification. IL bonds provide a type of asset that reduces the risks associated with inflation allowing investors to lock in a real rate of return.5 The demand of IL bonds has been largely driven by institutional investors, such as pension funds and insurance companies. The strategic demand for IL bonds is expected to continue growing as pension funds and other long-term funds increasingly use IL bonds as a long-term inflation hedge to shield their exposure to the long term nature of their liabilities, often also linked to inflation. In addition, the correlation of IL bond returns with those of conventional bonds and equities have been low or negative in most markets, making it possible for investors to improve the efficiency of their portfolios.

7. In contrast to the rapid expansion of global sovereign IL bond markets, UF-denominated debt issued by the Central Bank of Chile has seen a sustained decline in recent years. Until 2000, the majority of the central bank paper was denominated in UF, with peso-denominated and dollar-indexed instruments representing less than 20 percent of the total (Figure 2).6 With the shift to the nominalization of the monetary policy target, the central bank also adjusted its debt program to allow a sharp increase in nominal debt, with most of the increase reflecting the replacement of UF short term bills. In addition, the limited pre-announced periods of foreign exchange intervention in 2001 and 2002 contributed to higher issuance of dollar-linked debt. As a result, the share of UF-denominated paper fell to less than 45 percent by end 2003, with the stock of UF-denominated paper dropping to 60 percent of its capitalization in 2000.

Figure 2.
Figure 2.

Central Bank Paper by Currency, 1995-2004

Stock of paper (evaluated at end of period exchange rates)

Citation: IMF Staff Country Reports 2004, 292; 10.5089/9781451807608.002.A003

Source: Central Bank of Chile.

8. The drop in the stock of UF-denominated debt also reflected a change in the type of instruments supplied to the market. In 2002, the central bank announced a new debt program for medium- and long-term instruments, with a view to standardizing and increasing the efficiency of the UF market.7 The supply of the UF-linked promissory notes (PRCs), which had been issued for more than two decades, was discontinued and replaced with a more standardized UF-denominated bond program offering defined and sizeable bullets with 5-year and 10-year maturities (BCUs). In addition, the central government started issuing domestically 20-year UF-denominated securities in November 2003.

9. As in other countries, institutional investors in Chile have been the main drivers of demand for UF-denominated paper. The demand for UF bonds in Chile has been largely concentrated among pension funds and insurance companies, which held more than half of the outstanding stock. Despite increased liberalization of investment rules of these funds, strategic demand for UF-denominated instruments by the pension funds and insurance companies is likely to continue given the expected growth of these funds over the medium term and their continuous need to hedge their long-term inflation exposures.8

Some features of inflation-linked bonds

10. Although the structure and mechanics of IL bonds are similar across countries, differences do occur in some aspects of the bonds. The similarity in the structure of IL bonds, which reflects the fact that the real rate of return of these instruments must be known and fixed in advance, facilitates investors' understanding and pricing of these instruments across markets. This feature makes this type of bonds the only instrument for which income flows are fully adjusted for changes in the cost of living. However, differences do exist across markets:9

  • Deflation protection: Australia, France, and the United States offer a floor protection at par for the principal, whereas Chile and other markets do not.

  • Inflation lag: All bond programs are linked to inflation with a lag, allowing time for the compilation of inflation statistics. In Chile, the adjustment lag is one month with the UF allowing a lagged daily interpolation of the monthly inflation. In comparison, in Canada, France, and the United States, this lag is three months, and in the United Kingdom eight months.

  • Inflation index: As in Canada, the United Kingdom, and the United States, Chile uses a non-seasonally-adjusted headline CPI for inflation adjustment, while France uses a CPI excluding tobacco prices.

  • Taxation: Coupon income and principal appreciation are taxed as interest for most of the major issuing countries. However, in Chile, like in the United Kingdom, interests are taxed after adjusting for inflation. In France, principal appreciation is taxed as interest on an actuarial, smoothed basis.

  • Coupon frequency: Like in Canada, the United Kingdom, and the United States, IL bonds in Chile pay coupons on a semi-annual basis, while Australia and France do so on a quarterly and annual basis, respectively.

Linkages across IL bond markets

11. The growth of sovereign IL bonds has led to increased linkages between markets. Numerous studies have analyzed the extent to which short-term real interest rates are related across countries to evaluate the degree of financial integration. Cumby and Mishkin (1986) find that short-term real interest rates in several European countries and Canada are associated with U.S. real rate movements. However, Throop (1994) shows that the short-run responses between short- and long-term real rates are weak despite growing financial integration among industrial countries. In principle, international financial integration would not completely work to equalize real interest rates, as exchange rate expectations and time-varying risk premia may prevent convergence of real rates. Nonetheless, assuming constant ex-ante real exchange rates, higher integration would lead to higher co-movement of real rates in the long run.

12. The current analysis uses real yields from IL bonds with long maturities to assess the extent to which the Chilean UF bond market interacts with global IL bond markets. The data used includes weekly yields on IL bonds with maturities ranging from 8 to 18 years collected for Australia, Canada, Chile, France, South Africa, Sweden, the United Kingdom and the United States for the period from January 1, 1999 to April 30, 2004 (Figure 3).10

Figure 3.
Figure 3.

International Inflation-Linked Bond Yields

Citation: IMF Staff Country Reports 2004, 292; 10.5089/9781451807608.002.A003

Sources: Barclays Capital and RiskAmerica.

Since 1999, long-term IL bond yields have followed a downward trend in all the countries in the sample. The downward adjustment in yields reflected largely the global slowdown starting in 2000 and the concomitant easing of monetary policy. In fact, yields for all countries appear to follow unit root processes in the sample period. The analysis uses first differences of yields to assess the extent to which these series are integrated.

13. Granger-causality tests for yields across markets indicate precedence and information content of some markets. Table 1 summarizes the results from pairwise Granger-causality tests for all countries. A key finding is that IL yields in Australia, Canada, France, and Sweden are Granger-caused by three or more markets. In contrast, Chile, South Africa, the United Kingdom, and the United States do no seem to be influenced by other markets. An important difference among these countries, however, is that the United States and the United Kingdom do Granger-cause other markets. The United States in fact Granger-causes all other markets, except for Chile and South Africa. The little forecast value of the Chilean and South African yields on other yields underscores the relatively minor interaction of these markets with other markets. A VAR analysis using the system of yields across markets is also consistent with these findings.

Table 1.

Pairwise Granger Causality Tests of IL Bond Yields Across Markets

article image
Sources: Barclays Capital, RiskAmerica, and staff estimatesNote: (**) denotes significance at the 5 percent, and (*) at the 10 percent. Sample period Jan. 95–Mar. 04.Significance implies rejection of the null hypothesis meaning that row i does Granger Cause column j.

14. Impulse responses to U.S. IL bond innovations illustrate the dominant role of U.S. yields (Figure 4). Positive one-percent-deviation innovations of U.S. rates induce a positive response on most markets. For Australia, France, Sweden, and the United Kingdom, the innovation effect is significant even though the effect is short-lived. Surprisingly, the effect is less pronounced for the case of Canada despite the proximity of the Canadian and U.S. markets. Consistent with the previous evidence, innovations of U.S. yields are found to have little effect on Chilean and South African yields. Innovations of other markets do not appear to have any significant effect on these two markets either. Similar results are found after controlling for exchange rate movements and adjusting for country risk premia. This evidence would suggest that the Chilean and South African yields would appear to be less integrated to global IL bond markets. Still, market yields could also reflect, more fundamentally, differences in Chile’s economic and financial structure and on sources of economic fluctuations relative to those of typical advanced countries.

Figure 4.
Figure 4.

Impulse Responses to U.S. Yield Innovations

Citation: IMF Staff Country Reports 2004, 292; 10.5089/9781451807608.002.A003

Source: Fund staff estimates.

B. The Chilean Real Yield Curve as a Predictor of Output Growth

15. The nominal yield curve has been shown to contain information about real economic activity. It is well documented that the behavior of the nominal yield curve changes across the business cycle.11 The intuition follows from the fact that the premia on long bonds are countercyclical because of investors' dislike for risk in bad times, while shortterm yields are procyclical because of the stimulative monetary policy stance. In fact, the spread between nominal long-term and short-term government bond rates appears frequently in the literature as a significant regressor in equations that predict measures of future economic activity, with the predictive relationships robust over time and across different countries. The existence of a real yield curve in Chile provides thus a useful case to assess market expectations about future economic activity.12

Data and basic model

16. The study uses yield data from UF-denominated debt (PRCs) issued by the Central Bank of Chile with constant maturities ranging from 1–20 years spanning from January 1995 to March 2004. Real yields at various maturities, as measured from the yields of UF-linked central bank instruments with constant maturity, followed a common pattern since 1995 (Figure 5). From 1995 through early 1998, they remained within the six to eight percent range. In the fall of 1998, however, yields rose sharply as the central bank increased short-term rates to over nine percent to fend off pressure on the exchange rate. After 1999, yields at all maturities have experienced a gradual decline, with volatility in real yields rising sharply, especially in the short range, after September 2001 when the central bank moved to a nominal target rate.

Figure 5.
Figure 5.

Yields of UF-linked Bond (PRCs)

In percent

Citation: IMF Staff Country Reports 2004, 292; 10.5089/9781451807608.002.A003

Source: RiskAmerica.

17. Real yield spreads have followed a path consistent with the course of economic activity in Chile (Figure 6). From 1995 to 1998, the economy maintained its rapid growth pace observed since the mid-1980s. The spread between short-term and long-term UF-denominated interest rates was close to zero reflecting high short-term rates as a result of the prudent monetary policy stance. In fact, the real yield curve remained slightly inverted throughout the period with investors' demanding little long-term risk premia. After the monetary tightening in late 1998, real yield spreads fell sharply. The slowdown in economic activity and subsequent recovery in 1999 and 2000 followed the upturn in yield spreads. By the end of 2002, the slope of the curve had steepened significantly as monetary policy became increasingly loose and economic activity regained traction. In recent years, real yields on long-term bonds have hovered at around three to four percent and those on short-term yields at about one to two percent. The slope of the yield curve has been more volatile since 2001, reflecting the higher volatility of shortterm real yields associated with higher inflation volatility.13

Figure 6.
Figure 6.

Real Yield Spreads and Economic Activity

Real yield spreads

Citation: IMF Staff Country Reports 2004, 292; 10.5089/9781451807608.002.A003

Sources: Haver Analytics and RiskAmerica.

18. A simple model of the predictive power of the yield spread for economic activity can be formalized in the following regression:

gtk,n=αk+βk,nspreadtn+εtk,n

where future economic growth for the next k months is regressed on the term spread between the n-year maturity and the two-year maturity. The analysis uses three-, six-, and 12-month ahead growth horizons. Given that overlapping periods are used, the estimation uses a moving average correction to solve for serial correlation in the residuals. Real economic activity is measured by the seasonally-adjusted monthly indicator of economic activity (IMACEC). Table 2 reports the results of the regressions over the period January 1995–March 2004. All regressions have been adjusted to solve for heteroskedasticity and autocorrelation.

Table 2.

Forecast of GDP Growth in the Near Term from Yield Spreads

article image
Source: Staff estimates using data from RiskAmerica.com and Haver Analytics.Notes: (*) and (**) denotes significance at the 10 and 5 percent level, respectivelyAll regressions are corrected for heteroskedasticity and autocorrelation using the Newey-West method.

19. The results in Table 2 show that the term spreads at various maturities can help predict GDP growth in the near term. This finding is consistent with various ranges of the yield curve. Panel A uses medium-term spreads, as measured by the difference of yields between PRCs maturing in two and eight years. The coefficients of yield spreads on these regressions are all positive indicating that a positively sloped yield curve implies a pick up in growth, or conversely, a downward yield curve foretells a slowdown in activity. The relationship is significant when predicting six- and 12-month ahead growth.

20. The longer-term yields do not appear to improve the predictive ability of term spreads. A common finding in the literature using data for other countries suggests that the long maturity segment of the yield curve may lead to more efficient and accurate forecasts of GDP. In the case of Chile, however, using 18-year yields does not improve the fit of the regression, with the smaller, but still significant, coefficients of the yield spread. Nonetheless, the shorter range of the UF yield curve appears to be less successful in helping predict future GDP growth. None of the coefficients of the yield spreads in Panel B is significant for the entire period.

21. Not surprisingly, the ability of term spreads to anticipate GDP growth, however, appears to have diminished following the nominalization of the monetary policy target. The regressions using the January 1995-August 2001 subsample show more positive results. The size and significance of the coefficients of the yield spreads in the regressions increase noticeably with shorter range of UF yield becoming significant for the one-, six-, and 12-month ahead GDP growth horizon. The higher volatility of short-term UF interest rates in the aftermath of the nominalization of the monetary policy target appear to have clouded the information content of forward yields about expectations of economic activity in the near term.

22. For longer horizon forecasts, the predictive ability of yield spreads is less robust. The slope of the curve does not appear to be significantly correlated with output growth beyond the twelve-month ahead horizon (Table 3, Panel A). Interestingly, the long-term spreads measured by the difference between eight and 18-year yields is negatively correlated with 2-4 years ahead GDP growth. A negatively sloping curve in long range would be consistent with an expected business cycle recovery two to three years ahead.

Table 3.

Chile: Forecast of GDP Growth from Yield Spreads

article image
Source: Staff estimates using data from RiskAmerica.com and Haver Analytics.Notes: (*) and (**) denotes significance at the 10 and 5 percent level, respectivelyAll regressions are corrected for heteroskedasticity and autocorrelation using the Newey-West method.

23. Simple predictive regressions of future economic activity assume the exogeneity of yield spreads with regard to future economic activity. A VAR specification modeling the interaction between yield spreads and GDP growth shows that GDP growth does not Granger-cause changes in the slope of the yield curve. On the other hand, high yield spreads Granger-cause future GDP growth consistent with the univariate regressions. Hence, yield processes and future GDP growth do not appear to be jointly endogenous.

24. These findings underscore the forward-looking nature of the market information embedded in the yield curve and its ability to anticipate future economic growth. The entire yield curve has some predictive ability over the short run, in particular in the sub-period prior to the nominalization of the central bank monetary target. The predictive power of the real yield curve remains, however, low, especially after 2001. Out-of-sample forecasts using the simple model show the limited power to predict the magnitude of GDP growth at longer horizons.

C. The Role of Inflation-Linked Bonds in the Chilean Bond Market

25. The domestic corporate bond market has seen a remarkable growth in the past few years. Private sector bond issuance has doubled since the late 1990s, contributing to a fivefold increase in total outstanding debt stock from 1999 to 2003. Despite the central bank’s shift to nominalization and increased focus on nominal debt issuance, the corporate bond market remains, however, largely concentrated on the UF segment with most of the bond issues denominated in UF. Corporate debt is also characterized by its long duration, with the average maturity of bonds issued of about 14 years.

26. Several factors explain the preference for long-term UF-denominated paper. A key driver of the market is the large presence of institutional investors representing the main source of demand of corporate bonds, with pension funds and insurance companies having a natural preference for UF-denominated paper. Corporate firms have sought to cater to institutional investors, with most bond issuances in two tranches, one of 8–12 years, targeted to pension funds, and the other of 20 years, targeted to insurance companies.

27. Cost savings regarding inflation risk premium could remain another reason in favor of UF bonds. While inflation has seen a significant drop in the past decade and inflation expectations are well anchored within the central bank’s inflation target range, inflation volatility is still above that of more developed economies, suggesting that inflation risk premium could remain a significant cost. For economies with low inflation such as the United States and the United Kingdom, estimates of inflation risk premium suggest a cost of around 50 bps.14 Nonetheless, the cost savings from inflation risk premium in Chile is less evident when considering that the actual inflation premium has been consistently below most indicators of expected inflation.15

28. The existence of benchmark UF-denominated paper issued by the central bank has also been a critical factor for the development of the corporate bond market. Walker (2002) underscores the role played by the economic authorities in the development of the bond market with the issuance of indexed securities with diverse maturities, noting that the liquidity and benchmarking that these bonds provide are necessary signals for the private sector. In particular, the central bank sought to promote longer-term bonds with more than a third of its instruments with maturities five years or higher. In addition, while the average maturity of the central bank paper in UF has fallen since 2001 (with less than 15 percent of outstanding UF paper at end 2003 having maturities five years or higher), the 20-year bonds first issued by the central bank (BCUs) and recently issued by the central government (BTUs) have longer duration than previously issued central bank bonds and have served as a reference for longterm placements of corporate bonds.

29. The shift to a nominal monetary policy target and recent changes in the central bank’s debt program does not appear to have had a significant impact on the volatility of corporate bond yields. Specifically, the higher volatility in benchmark short-term yields did not translate into higher volatility for long-term corporate bond yields (Figure 7). Short-term yields, however, have shown increased volatility in recent years, similar to that observed on central bank short-term bonds. Volatilities of yields across risk categories were also largely unaffected. In fact, since the implementation of the central bank’s new bond program, there has been improved liquidity in the markets, providing market price signaling for private issuers.

Figure 7.
Figure 7.

Corporate Bond Yields

Citation: IMF Staff Country Reports 2004, 292; 10.5089/9781451807608.002.A003

Source: Central Bank of Chile.

30. A key challenge for corporate financial officers remains the diversification of their liabilities toward peso debt. The central bank has aggressively expanded the issuance of short- and medium-term peso debt. Indeed, the issuance of two- and five-year maturity peso bonds accounted for 40 percent of the central bank’s total primary offerings. The recent announcement of the new 10-year bond will probably help complete the market of peso bonds and provide an important stimulus to the issuance of corporate peso bonds. In this regard, long-term nominal bonds could have a role, not only in terms of clarifying private sector expectations about future inflation rates, but also to facilitate the participation of foreign investors that might perceive nominal bonds as a simpler instrument to invest in.

D. Concluding Remarks

31. This chapter has reviewed the role of UF-denominated paper in the central bank’s debt program. The analysis compared the development of the UF bond market to the rapid growth of the global market of inflation-linked bonds. In particular, IL bond yields across markets show some degree of correlation and variance decomposition indicates the presence of common factors. Granger-causality tests indicate that the U.S. market plays a leading role in global IL bond markets. Chilean UF yields, however, would appear to have less interaction with other markets, suggesting that Chilean domestic markets could also reflect differences in Chile’s economic structure and on sources of economic fluctuations relative to those of typical advanced countries. In this regard, Chilean bonds could potentially provide good diversification value.

32. The analysis considers one of the ancillary merits of maintaining a well-functioning UF bond market pointing to the value of the real yield curve as a tool for policy makers to assess market expectations about future economic activity. The main result shows the ability of the real yield curve to anticipate future economic activity, underscoring the forward-looking nature of the information embedded in the yield curve. In particular, spreads between two- and eight-year maturity UF bonds are strongly related to the 12-months ahead GDP growth. However, with the nominalization of the monetary policy target, the predictive power of the UF yield curve has diminished, as short-term UF yields have been more affected by high frequency noise of inflation dynamics.

33. Finally, the chapter underscores the role of the central bank’s UF-denominated paper in the development of the corporate bond market. The shift to nominalization of the monetary policy target rate does not appear to have had an impact on the volatility of UF yields. The higher standardization of central bank issuance since September 2002 has facilitated price discovery in the corporate sector issuance by boosting the liquidity and size of benchmark bonds. The simplification and standardization of the interest term structure could help promote foreign investors' participation in the Chilean domestic market. In addition, the central bank’s efforts to extend the maturity range of the peso yield curve could provide the appropriate stimulus for the development of a corporate peso-denominated bond market.

References

  • Bank of Canada, 2003, “Debt Strategy 2004-2005,Consultation Review, December.

  • Barclays Capital, 2003, “Barclays Inflation-inked Bond Indices”, Barclays Capital Research Index Products, September.

  • Bodie, Z., 1990, “Inflation, Index-Linked Bonds, and Asset Allocation,Journal of Portfolio Management, Vol. 16, No. 2, pp. 4853.

    • Search Google Scholar
    • Export Citation
  • Bridgewater Associates, 2002, “Inflation Linked Bond: Observations,” (unpublished, Bridgewater Associates).

  • Campbell, J. Y., and R. J. Shiller, 1996, “A Scorecard for Indexed Government Debt,” in NBER Macroeconomics Annual 1996, edited by B. Bernanke and J. Rotenberg, pp. 15597, Cambridge, Massachusetts: MIT Press.

    • Search Google Scholar
    • Export Citation
  • Cumby, R., and F. Mishkin, 1986, “The International Linkage of Real Interest Rates: The European-U.S. Connection,Journal of International Money Credit and Finance, Vol. 5, pp. 523.

    • Search Google Scholar
    • Export Citation
  • Estrella, A., and F. Mishkin, 1997, “The Predictive Power of the Term Structure of Interest Rates in Europe and the United States: Implications for the European Central Bank,European Economic Review, Vol. 41, pp. 13751401.

    • Search Google Scholar
    • Export Citation
  • Hamilton, J., and D. Kim, 2002, “A Re-Examination of the Predictability of the Yield Spread for Real Economic Activity”, Journal of Money Credit and Banking, Vol. 34, pp. 34060.

    • Search Google Scholar
    • Export Citation
  • Landerretche, O., F. Lefort, and R. Valdés, 2002, “Causes and Consequences of Indexation: A Review of the Literature”, in Indexation, Inflation, and Monetary Policy edited by in Lefort, F. and K. Schmidt-Hebbel, (Santiago: Chile).

    • Search Google Scholar
    • Export Citation
  • Plosser, C., and G. Rouwenhorst, 1994, “International Term Structures and Real Economic Growth”, Journal of Monetary Economics, Vol. 33, pp. 13555.

    • Search Google Scholar
    • Export Citation
  • Shiller, R. J., 2002, “Indexed Units of Account: Theory and Assessment,” in Indexation, Inflation, and Monetary Policy edited by F. Lefort and K. Schmidt-Hebbel, (Santiago: Chile).

    • Search Google Scholar
    • Export Citation
  • Throop, A., 1994, “International Financial Market Integration and Linkages of National Interest Rates,Economic Review—Federal Reserve Bank of San Francisco, Vol. 3, pp. 318.

    • Search Google Scholar
    • Export Citation
  • Walker, E., 2002, “The Chilean Experience in Completing Markets with Financial Indexation,” in Indexation, Inflation and Monetary Policy edited by F. Lefort and K. Schmidt-Hebbel, (Santiago: Chile).

    • Search Google Scholar
    • Export Citation
1

Prepared by Rodolfo Luzio (Ext. 3-8327).

3

The trade-off between broadening financial markets and the risk of fragmenting market liquidity is an issue that has received significant attention in countries with falling or low debt stocks, such as Australia or Canada. See, for instance, Bank of Canada (2003).

4

Most recently, the U.S. Treasury announced an aggressive expansion of its IL bond program with expected net new issuance of about US$60–70 billion a year in the next couple of years, about a third of the current total market capitalization.

5

See Bridgewater (2002) for an efficient frontier analysis showing the degree to which IL bonds merit inclusion and that they tend to displace nominal bonds with a typical portfolio. Bodie (1990) shows how the introduction of IL bonds can improve portfolio efficiency, and why these instruments are the only hedge against long-run inflation risk.

6

This is in stark contrast to most other sovereign IL bond markets, which, despite their rapid growth, still represent a small share of the total sovereign debt markets.

7

Another important innovation included the stripping of coupons allowing the development of zerocoupon instruments.

8

Pension payments are tied to the UF and the minimum pension guarantee is also linked to the UF.

9

The econometric analysis below does not take into account the pricing implications of these minor structural differences.

10

Risk America.com kindly provided the UF yield data for Chile. The data for other countries was obtained from Barclays Capital.

12

The theoretical basis for the empirical evidence of the predictive power of the yield curve refers mostly to real yields rather than nominal yields. For instance, Plosser and Rouwenhorts (1994) use real business cycle models to explain the relationship between the term structure and real activity.

13

Following the nominalization of the monetary policy target, short-term UF yields moved closely with inflation as short-term nominal interest rates were anchored around the monetary policy target.

15

Indeed, breakeven inflation, measured as the difference between nominal and real yields of central bank paper, has been consistently below inflation expectation measure of the Central Bank of Chile’s survey of market participants and the Bank’s inflation projections.

Chile: Selected Issues
Author: International Monetary Fund