2003 Market Consultations on Real Return Bonds
Summary of Comments
The Government of Canada actively seeks input from participants on an ongoing basis on the performance of and potential adjustments to federal debt programs. Consultations are conducted regularly to help the Government meet its strategic objective of maintaining a well-functioning market for Government of Canada securities. The purpose of these consultations was to obtain views on issues relating to the design and operation of the Government of Canada's Real Return Bond (RRB) program.
In its Debt Management Strategy 2003-04 publication, the government announced a review of the RRB program. Consultations were subsequently held with a wide range of market participants in September and October 2003. The views obtained from these consultations are one of a number of considerations in the evaluation of the program.
During the consultations, the Department of Finance and the Bank of Canada received input from more than 50 institutional stakeholders, including Primary Dealers, Government Securities Distributors, institutional investors, and pension consultants. In addition to these participants, the government also received comments from institutional and individual (retail) investors via the consultation email address.
To facilitate the discussion of RRBs, the government published a questionnaire for participants. The questionnaire covered four major topics relating to RRBs:
- Use and treatment
- Current and future demand
- Demand for substitutes
- Development of primary and secondary markets
Overall, market participants were of the view that RRBs provide value and that the government should continue to issue them. Participants suggested that current and future demand for RRBs would be robust mainly because of strategic investment by firms and individuals with inflation-linked liabilities and because of the diversification benefits that RRBs provide. Participants generally found that the primary market functions well and that conditions in the secondary market have improved, although liquidity is naturally lower in the secondary market than in the nominal bond market, given the buy-and-hold nature of most RRB investments. There was no consensus on how to improve the product's design or on improving liquidity in the secondary market.
Use and Treatment
The vast majority of participants characterized RRBs as buy-and-hold instruments. The largest holders of RRBs, the domestic pension fund community, hold RRBs in their core portfolio to match long-term inflation-sensitive liabilities. A number of participants noted the risk-diversification benefits of RRBs: they allow fund managers to take on more risk in other asset classes. The average target allocation for accounts with formal RRB mandates varied in the 5-10 per cent range of their total portfolio.
Aside from their use as a strategic investment, certain domestic and international accounts actively manage tactical RRB portfolios.
Retail investors indicated that they hold RRBs as a source of income with protection against inflation upon retirement.
RRBs are not yet widely treated as a separate asset class from fixed-income securities, but acceptance of this view is growing. Investors who currently treat RRBs as a separate asset class are primarily those with liability-matching requirements, such as pension plans or endowment funds. These investors tend not to measure the performance of their RRB holdings relative to that of other asset classes or indexes since they are using RRBs to match liabilities.
With respect to duration and benchmark measurements, there is no consistency in the methods used by participants. The most widely used external benchmarks are Scotia Capital's Canadian Bond Universe and RRB indexes. Some participants also noted that RRBs provide a good proxy for the risk-free asset in the relative evaluation of equities and nominal bonds and for comparing the Canadian-dollar pricing of bonds with foreign bond market valuations.
Current and Future Demand
The strong demand experienced in the recent past was attributed to increasing strategic and tactical decisions regarding asset allocation, based on a better understanding of the role of RRBs in investment portfolios and their behaviour in the secondary market. Pension fund consultants have been recommending that institutional investors hold RRBs in their core portfolios for asset-liability management and risk-minimization purposes. This has led some institutional funds to build up sizable portfolios in order to reach the recommended or newly mandated RRB allotments.
Strategic demand for RRBs has also increased because of the relatively poor performance of equities until the spring of 2003 and the low yields in fixed-income markets in recent years. Some participants noted that this coincided with a change in the equity risk premium that caused a general shift from equities to fixed-income assets. In addition, since RRBs were one of the best-performing asset classes over the past few years, a number of pension plan sponsors have chosen to include RRBs in their investment policies.
Almost all participants agreed that, over the past year, the level of tactical activity has been high. International accounts have been among the most active, taking advantage of yield spreads between RRBs and the inflation-linked bonds (ILBs) of other sovereigns. Some commented that this activity reflects a greater global acceptance of ILBs, as well as increased ILB issuance globally by long-standing and new sovereign borrowers in the market.
Most participants expect aggregate demand for RRBs to increase. Market participants indicated that future demand for individual portfolios would be a function of various factors including RRB holdings relative to target allocations, portfolio rebalancing, coupon re-investment, and changes in investment mandate by plan sponsors. Managers of tactical portfolios generally said their future demand would be determined by market dynamics and by trading opportunities that may arise.
More fundamentally, it was noted that strategic demand is expected to grow as the mandates of pension funds, endowment funds, and foundations increasingly include RRBs for matching funds' liabilities. In the same vein, some participants suggested that part of the increase in strategic demand would also come from under-funded private pensions seeking to achieve a minimal rate of return. The natural growth of fund assets could also justify higher demand for RRBs through increasing coupon re-investment needs and portfolio rebalancing. Changing demographics stemming from the aging population should also increase the natural demand for RRBs.
There were mixed views on the extent of rollover activity in RRB holdings, where existing issues are switched for new, longer-dated maturities. Participants had mixed opinions on the potential use of switch buybacks for RRBs to address these rollover needs. Those in favour of buybacks mentioned that switches involving the ability to buy or sell benchmark RRBs for off-the-run RRBs (i.e., two-way switches) would be appropriate, although they would work best with RRBs of less than 10 years to maturity. Those accounts that typically buy and hold RRBs suggested that there may not be a pressing need to reduce the number of old RRB issues.
A number of dealers and fund managers noted increasing retail demand. Comments from retail investors indicated that RRBs are a useful financial instrument when preparing for retirement.
Demand for Substitutes
Most participants said that there is currently no perfect substitute for RRBs in the Canadian market. Participants are apprehensive about holding non-Government of Canada ILBs, given credit concerns and the long duration of the inflation-linked liabilities being matched. However, if RRBs were unavailable, participants cited a number of alternative investments including, provincial ILBs, corporate/infrastructure ILBs, and real estate. The general sentiment was that most alternative products, especially non-government RRBs, offered an imperfect hedge against inflation and had inherent credit and liquidity risks. The lack of supply and/or low liquidity in existing non-Government of Canada ILBs were particularly noted. Other substitute products that were less popular choices included commodities, equities, floating-rate notes, and t-bills.
The vast majority of funds do not hold any ILBs of other sovereign issuers as part of their core portfolio. The most frequent reason given for not holding foreign-denominated ILBs is that fund liabilities are in Canadian dollars and are a function of Canadian inflation. Another reason given for not holding foreign-denominated ILBs is the foreign property rule. However, some participants said that they would consider trading foreign ILBs on a tactical basis in the future.
Development of Primary and Secondary Markets
In general, institutional investors and intermediaries thought that the primary market functions well, particularly the Dutch auction process (single-price auction). According to some participants, Dutch auctions allow institutional investors to bid aggressively and decrease the likelihood of poor auction results for the government. There was mention of the potential benefits to institutional investors of direct access to auctions. Some participants also suggested that the government could even conduct issuance on a tap basis, but this sentiment was not widely held.
While some participants either felt that current auction sizes are sufficient or did not have a firm view on the matter, a number of dealers and accounts said that the government should increase auction sizes, citing excess demand. Higher demand is particularly prevalent at auctions around coupon payment dates because of demand for coupon reinvestment.
Some participants indicated that they would accept reductions in the supply of nominal bonds and treasury bills to permit a moderate increase in annual RRB issuance.
Some market participants thought the issuance of RRBs with maturities other than 30 years could be beneficial. In particular, a 10-year issue would extend the real yield curve, possibly making markets more efficient. There were mixed views, however, on where the potential demand for a 10-year issue would come from. The most frequent candidates mentioned were retail clients and tactical accounts. Conversely, many institutional clients stated that a 10-year product would be of little use to them in matching long-term liabilities. Several participants observed that the current stock of RRBs would eventually roll down and create a real yield curve.
Participants were generally indifferent to the coupon-setting process for RRBs. It was also suggested that there would be no value attributed to adding deflation protection on the RRB principal amount, since the probability of deflation is perceived to be low over the life of a 30-year RRB issue. A common suggestion from institutional investors was to expand the information released in auction results to include the aggregate allotment amounts between clients and dealers.
Participants generally agreed that liquidity in the secondary market has improved over time but remains lower than the level of liquidity found in the nominal bond market. A number of factors were cited, including the buy-and-hold nature of the investor base, lack of supply, and the absence of an effective hedge against RRBs available to dealers. Some participants commented that the buy-and-hold nature of demand was the determining factor that would likely keep liquidity constrained, regardless of issuance size.
Participants observed that the RRB market still cannot accommodate trades of significant size without a major impact on prices. Some mentioned that liquidity is sufficient if counterparties are willing to enter into multiple transactions of limited sizes over a certain time period. Typically, participants find trades up to $10 million not too difficult to transact. But larger trades of $25 million and above are generally more difficult to execute. Some dealers noted that they tend to steer large orders from tactical investors towards auctions, rather than trying to fill them in the secondary market. Some participants noted that liquidity in the secondary market peaks during the two-week period around each auction, especially on coupon payment dates, but otherwise tends to be sporadic.
Market participants stated several reasons why RRBs are sometimes more expensive for the government than nominal bonds. These include a liquidity premium because of the lower level of liquidity for RRBs compared with nominal bonds, the divergence of expected and actual inflation, and the absence of effective hedging instruments that may cause RRB yields to be higher at times.