INVESTMENT MANAGEMENT Australian Fixed income An alternative approach MAY 2017 macquarie.com
Important information For professional investors only not for distribution to retail investors. For recipients in Australia: This document is issued by Macquarie Investment Management Global Limited (ABN 90 086 159 060 AFSL 237843). The information in this document is provided for general information purposes only and is not, and should not be construed as, an advertisement, an invitation, an offer, a solicitation of an offer or a recommendation to participate in any investment strategy or take any other action, including to buy or sell any product or security or offer any banking or financial service or facility by any member of the Macquarie Group. This document has been prepared without taking into account any person s objectives, financial situation or needs. Recipients should not construe the contents of this document as financial, investment or other advice. It should not be relied on in making any investment decision. Future results are impossible to predict. This document contains opinions, conclusions, estimates and other forward-looking statements which are, by their very nature, subject to various risks and uncertainties. Actual events or results may differ materially, positively or negatively, from those reflected or contemplated in such forward-looking statements. Past performance information shown herein, is not a reliable indicator of future performance. No representation or warranty, express or implied, is made as to the suitability, accuracy, currency or completeness of the information, opinions and conclusions contained in this document. In preparing this document, reliance has been placed, without independent verification, on the accuracy and completeness of information available from external sources. To the maximum extent permitted by law, no member of the Macquarie Group nor its directors, employees or agents accept any liability for any loss arising from the use of this document, its contents or otherwise arising in connection with it. Other than Macquarie Bank Limited (MBL), none of the entities noted in this document are authorised deposit-taking institutions for the purposes of the Banking Act 1959 (Commonwealth of Australia). The obligations of these entities do not represent deposits or other liabilities of MBL. MBL does not guarantee or otherwise provide assurance in respect of the obligations of these entities, unless noted otherwise. 2
Contents Executive Summary 4 Introduction 5 The current environment 6 The reliability of enhanced strategies 6 The flaws with traditional active strategies 6 Increased alpha targets decrease diversification benefits 6 Credit risk is not the panacea 6 An alternative approach to managing active Australian fixed income 8 Arbitrage 8 Futures arbitrage 8 Covered Interest parity 9 Near-arbitrage/high reliability 10 Currency carry 10 Curve carry 10 Forward curve relative value 11 Better directional trading 11 Conclusions 14 3
About the author DEAN STEWART EXECUTIVE DIRECTOR FIXED INCOME AND CURRENCY Dean is primarily responsible for investment research and process development. In this role, he combines his long and varied fixed income experience with analytical rigour to continually improve our investment processes and product designs to ensure that maximum value is extracted from the available opportunities in fixed income markets. This is to ensure that we continue to deliver the exceptional performance clients have come to expect. During Dean s many years at Macquarie, he has held numerous senior roles, including portfolio management roles, Head of Credit, Head of Research, and Head of Fixed Income and Currency. As Head of Credit, he was the portfolio manager responsible for all credit investments in all of the interest rate funds managed by Macquarie. As Head of Research, Dean authored a number of research papers which have become the backbone of the investment philosophy and processes employed by the team, including An investigation into duration and Crisis management. Prior to this, Dean was Portfolio Manager for our enhanced fixed income solutions, took responsibility for portfolio construction across all fixed income funds and was responsible for trading duration in our credit enhanced cash funds. MATTHEW MULCAHY ASSOCIATE DIRECTOR SENIOR PORTFOLIO MANAGER FIXED INCOME AND CURRENCY Matthew co-manages Macquarie s Australian fixed interest portfolios. His primary responsibility is managing the duration positioning as well as portfolio construction. Matthew s experience contributes to the investment strategies across all the cash and fixed income solutions globally. Matthew has over 18 years industry experience, and has a Bachelor of Business (Accounting and Finance) from the University of Technology Sydney and is a CPA. 4
Executive summary Australian fixed income managers have often shown a good ability to add value in small amounts, but very poor consistency when targeting the types of excess returns expected from core and core plus strategies. This paper examines the evidence and reasons for this, and offers better alternatives for investors seeking core style excess returns. Several examples are given of strategies that deliver arbitrage or near arbitrage return opportunities. Our findings indicate that the most common approach for core fixed income managers to achieve higher target returns in core and core plus strategies is to scale up existing strategies. However, this tends lead to an undesirable outcome as taking larger positions in more volatile strategies (such as duration trading) increases the volatility of returns and can destroy the alpha already achieved. This is largely attributable to factors such as; limits on gearing, restrictions on short selling and the interaction effects from block positioning contributing to the inability to add alpha by simply scaling up existing strategies. Another strategy employed to increase returns is to utilise allocations to credit risk. This strategy is also flawed as it promotes a chase for yield mentality where investors look for yield at the expense of the risk entailed this is particularly concerning given the asymmetric profile of credit risk whereby losses from downgrades far outweigh the gain from upgrades. However, through the incorporation of a large number of diverse, low-risk alpha strategies, it is possible to increase alpha without compromising consistency of returns. By adding many high information ratio strategies including, among others, futures arbitrage, covered interest parity and directional duration trading, we demonstrate the potential to manage core and core plus strategies using a much better approach. 5
Introduction Many Australian fixed income managers have had a tendency in the past to overstretch. While research shows that Australian fixed income managers have the ability to generate returns consistently above the benchmark through enhancement strategies, when they try to scale up the alpha they can generate to higher levels they tend to destroy it. The chart below demonstrates this. It shows the long term excess returns that Australian fixed income managers have delivered at different levels of risk. These numbers have been adjusted for survivorship bias, as detailed in the paper Unconstrained Funds: The right strategy to deal with low yields? Chart 1: Risk and Excess Return for Different Australian Fixed Income Styles 1.00% 0.50% Excess Return 0.00% -0.50% -1.00% -1.50% 0.00% 0.50% 1.00% 1.50% 2.00% Excess Return Standard Deviation Core Core Plus Enhanced Source: Macquarie and Mercer This causes a conundrum; everyone wants more alpha, but reaching too far consistently destroys what was already achieved. However, there is a better way to increase the alpha in an Australian fixed income portfolio; one that doesn t rely on taking large, volatile, duration positions. The approach requires building on the strengths of an enhanced 1 style, then broadening it to include many and varied directional interest rates strategies rather than just being long or short duration. 1 Not all enhanced styles are the same! In particular, credit enhanced strategies perform extremely poorly in crisis situations. For research into the robustness of different styles of enhanced fixed income, see the Macquarie Investment Management paper, Seeking to Enhance Fixed Income without Enhancing Losses, 2009. 6
The current environment The reliability of enhanced strategies Enhanced funds generally achieve their returns through a vast number of small security selection decisions in government, semi-government and corporate bonds, sometimes augmented with very low risk yield curve, duration, and sector rotation strategies. The important thing about the types of positions in enhanced fixed income funds is that they are small and numerous. Using many small positions leads to consistency of overall alpha because the fund benefits very significantly from diversification. For example, if a fund has a single strategy, and that strategy has a 55% probability of adding value, then the fund will add value with 55% probability. On the other hand, if the fund has 200 independent and equal strategies, each with 55% probability of adding value, then the fund will add value with approximately 90% 2 probability. The best enhanced funds do add value in approximately 90% of months. The flaws with traditional active strategies Increased alpha targets decrease diversification benefits Unfortunately, when alpha targets are increased and alpha sources scaled up, as they are in core and core plus funds, the diversity of positions is reduced, and therefore, so is the diversification benefit. There are several reasons for this: 1 Limits on gearing Some positions, while highly reliable, rely on very small moves in pricing, and therefore need large exposures to have an impact on performance. The return on these positions cannot be increased without gearing the portfolio. Funds are usually not permitted to gear, and even if they are, the cost of funding leveraged positions can eliminate the alpha they hope to exploit. In fact, the alpha is sometimes available precisely because the hedge funds and sell-side firms that usually take advantage of these opportunities are unable to do so due to the cost of funding. 2 Restrictions on short selling Other alpha generation strategies rely on selling holdings in order to be underweight relative to a benchmark. Once all the holdings in a security are sold, the position cannot be scaled up without short-selling a security. Again, funds are usually not permitted to short-sell bonds, and even if they are, the cost of funding short-sales can offset the opportunity. 3 Interaction effects block positioning Positions can interact with each other. For example, it is difficult for a fund to have a yield curve steepening position and also be long duration, because the yield curve steepening position involves buying short maturity bonds and selling long maturity bonds, which reduces duration. In practice therefore, a fixed income portfolio manager can t scale up alpha simply by taking the same positions in larger sizes across the board. As a result, portfolio managers typically take larger positions in more volatile strategies, such as duration trading, reducing consistency of returns. This may also restrict the managers from benefiting from the low volatility positions that an enhanced fund would, further reducing the consistency of alpha. Credit risk is not the panacea A further issue occurs when trying to scale up risk in credit holdings. Managers looking to add a lot of value through credit will reach a point where they can t simply buy more of the same credits as they would with more conservative targets, but must buy different ones. They will buy the types of securities that have wider credit spreads than the market; securities with higher risk. Often, due to credit constraints, these will be the securities have the worst fundamentals in their rating band; the securities most likely to be downgraded. Holding a portfolio of such securities will lead to underperformance for a couple of reasons. 2 Assuming each strategy either makes or losses the same amount, this probability can be computed from the binomial distribution with 100 successes in 200 trials and 55% success rate. 7
Firstly, other managers will likely be targeting the same securities; not because they offer good value for their credit risk, but because they offer high yield for their rating. We could say, they offer high credit risk for their rating. Managers can be pressured to buy these securities even when they offer poor value for risk because they are trying to push their returns within fixed constraints. For this reason, securities that are poor credit quality for a particular rating often don t reflect a fair price for the risk involved. Secondly, this approach is philosophically flawed. As Macquarie has written about extensively in the past 3, downgrades occur more than upgrades, and the loss from a downgrade is larger than the gain from an upgrade, so the best way to add value with a credit portfolio is to avoid the downgrades, rather than to chase additional yield (from securities more likely to be downgraded). Combining the issues of scaling up credit returns with the issues of scaling up relative value explains why core funds typically have much lower consistency than enhanced funds. Is an investor looking for a fixed income portfolio therefore constrained to have a high probability of low alpha from an enhanced fund or a low probability of higher alpha from a core or core plus fund? If core funds use the traditional approach? Yes, investors face a choice between higher alpha and lower consistency or lower alpha and higher consistency. 3 For Macquarie Investment Management s latest paper on managing corporate credit risk where we discuss this fact see The changed nature of credit investment, 2015. The first of this series of papers was written in 1997. 8
An alternative approach to managing active Australian fixed income However, there is a better way; a third choice with higher alpha and higher consistency. Instead of sacrificing the consistency of a large number of small positions, replacing them with a small number of large positions in a search for higher alpha, it is possible to use an even larger number of small positions. Small security selection strategies can be augmented by a diverse array of directional and curve strategies, that are different to traditional duration and yield curve trading. Some of these directional and curve strategies include: Arbitrage Futures arbitrage An arbitrage trade is defined as a trade that produces a riskless profit. Textbooks will often say such opportunities shouldn t exist or would only be fleeting, but they assume that everyone has the same information, the same motivation, the same objectives, and that markets are frictionless. In reality, there are many arbitrage opportunities that exist for sustained periods of time because different people have different objectives. For example, if a combination of Australian bond futures and bank bills provides a different return to a physical bond, there is an arbitrage opportunity to sell one and buy the other. There is nothing new in using arbitrage positions in Australian fixed income funds; enhanced funds often use them. The extent of their use can be extended in active funds to take similar positions in different futures contracts across the curve, and in different futures markets around the world, creating scores of opportunities for futures arbitrage. Arbitrage trades can also be done by selling futures, buying the bond then placing it on repo. The chart below shows how the arbitrage pricing between bonds and the Australian 3 year bond future changes over time. Whenever the blue line is not at zero, an opportunity exists for riskless profit. Chart 2: Australian Three Year Futures Arbitrage 4 2 Futures cheap Futures excess yield (bps) 0-2 -4-6 Futures expensive -8 Oct 15 Jan 16 Apr 16 Jul 16 3 year bond vs futures price Source: Bloomberg 9
Because of the volatility of the arbitrage, the profitability of the strategy also has volatility. Although the profit is known on a quarterly basis when futures roll, profitability varies during the quarter. The profitability of this strategy as managed by Macquarie has an information ratio of 1.59. Covered Interest parity According to covered interest parity, investing in offshore cash and hedging it back to Australian dollars should achieve the same return as investing in Australian cash. However, covered interest parity doesn t always hold at all maturities along the bill curve and in all countries (and sometimes it doesn t hold at any maturity). In these cases, it is possible to invest in offshore bills, hedge the currency out, and achieve higher Australian dollar bill returns than available in the local market at the same credit quality. With dozens of highly rated liquid markets around the world, the opportunities here are extensive. The chart below shows the additional yield available from investing in offshore bill securities, and hedging the currency back to Australian dollars. The countries are real; the names have just been withheld. The excess yield is essentially a riskless profit. Chart 3: Offshore LIBOR excess yield (Hedged to AUD) 0.8% 0.6% Excess yield 0.4% 0.2% 0.0% -0.2% Country A Country B Country C Country D Country E Country F 3 Months 6 Months 12 Months Source: Bloomberg and Macquarie For the past year, investing in the 3 month instruments of Country A, currency hedged produced an excess return of 50 basis points with an information ratio of 2.9. Investing in the 12 months instruments produced an excess return of 66 basis points with an information ratio of 2.1. 10
Near-arbitrage/high reliability There are other positions that managers can take that are not pure arbitrage opportunities, but have low risk and high reliability. Examples include: Currency carry The higher interest earned from investing in a country with high interest rates should be offset by an associated depreciation in that currency. However, the depreciation in high interest rate currencies is often less than would be expected from covered interest parity. Indeed, often the currencies with higher interest rates can actually increase in value as people seek higher yields. By taking small overweight positions in a number of high carry currencies, and small underweight positions in a number of low carry currencies, the returns of a fixed income portfolio can often be improved. This can only be done in small amounts, and only by teams with currency management capabilities and expertise because at times the return from this style of carry can correct quite suddenly but in small amounts, over time, currency carry can contribute to returns. Curve carry Long dated bonds normally have higher yields than short dated bonds, and so investors pick up additional yield (carry) from holding longer duration bonds. In different countries, with different sloped yield curves, this form of carry can vary. Because there are many different futures available globally (and an almost limitless availability of swaps and bonds), it is possible to increase the carry of a portfolio with very low risk. An important characteristic of curve carry is that it is negatively correlated to other forms of carry, such as currency and credit carry. Therefore it can prove very helpful in increasing the overall reliability of the portfolio. The chart below shows the additional return that has been delivered by a specifically structured curve carry strategy over a number of markets over the past 5 years. These results are live out-of-sample results, not a back-test. The information ratio on this strategy is just over 1.1 which equates to steady consistent additional value as shown in the chart below. Chart 4: Cumulative hedged global carry strategy return 5.0% 4.5% 4.0% 3.5% Cumulative return 3.0% 2.5% 2.0% 1.5% 1.0% 0.5% 0.0% 2011 2012 2013 2014 2015 2016 Source: Macquarie 11
Forward curve relative value While the ability to extract value from relative value trades in bonds is often limited by the amount of the bond available, relative value trades in swaps and forward curves are far more scalable, and a more precise method of targeting opportunities. Small kinks in a swap curve can imply very significant differences in forward rates, sometimes leaving portions of the forward curve with irrationally high or low yields. Portfolio managers can take advantage of this by receiving the parts of the forward curve with very high yields, and paying the parts of the curve with very low yields. It is also possible to do relative value trades between the physical bond curve and the forward curves. When participating in these types of trades it is important to understand that the demand for different types of instruments (bonds, futures, forwards and swaps) is driven by different investors, and to understand the flows and motivations of these investors. The table below shows how forward rates can vary a lot by changing maturities by only a small amount, offering opportunities for alpha. The expensive areas are shaded in red and the cheap areas in green. The same opportunities present themselves in many countries. Table 1: Sample of Australian Forward pricing matrix August 2016 Forward periods 3 months 6 months 1 year 2 years 3 years 4 years 5 years 6 years 7 years 8 years 9 years 10 years 11 years Maturities 1 1.81 1.79 1.79 1.84 1.92 2.00 2.13 2.28 2.42 2.38 2.55 2.51 2.64 2 1.80 1.80 1.82 1.88 1.96 2.06 2.21 2.35 2.40 2.46 2.53 2.55 2.66 3 1.82 1.83 1.85 1.92 2.02 2.14 2.28 2.36 2.45 2.48 2.55 2.58 2.68 4 1.85 1.86 1.89 1.97 2.08 2.21 2.30 2.40 2.46 2.50 2.57 2.62 2.70 Source: Bloomberg and Macquarie In the Australian market alone over the past 12 months, this strategy delivered an information ratio of 2.4. Better directional trading Trading the whole distribution, not just the direction Many portfolio managers execute duration positions in a very simple manner. They decide whether to be long or short duration relative to the benchmark, and then trade to position the portfolios there. This is a difficult way to add value because there are thousands of other investors also taking a view on whether the current interest rate will rise or fall, bringing the probability of getting it right to roughly 50%. For a market in equilibrium, approximately half the market will expect the market to go up and half expect it to go down. However, it may be that those who believe it will go up only think it should rise a small amount, and will sell if it does. In addition, the people who originally thought yields should rise may buy more bonds because of better levels. This creates bulges in the distribution of future bond yields that can be used to create alpha. However, most people only take a view on the one part of the distribution represented by current rates. An example of potential bulges in the distribution of future interest rates can be seen by looking at the distribution of past rates. There are discontinuities in the distribution of past movements that align with discontinuities in the distribution of future trading levels. 12
This can be better demonstrated by looking at an example of the distribution of past trading levels at a fixed point in time for a specific past time-frame, as shown below. At this particular point in time, without new information in the market, future levels will have a tendency to follow the distribution of the right-hand peak of the distribution (that is, the one closest to current levels). Should the market trade near the right hand side of the distribution, it is more likely to fall than rise. Should it trade near the left hand side, it is more likely to rise than fall. However, if the trading level moves past the peak in this distribution (at -0.08) it may continue through the area of little trading until the previous peak at -0.20. Options can be used to position the portfolio to make profits when the market trades within this range of higher probability. Chart 5: Distribution of 3 year Australian bond futures trades (4 months to May 2016) 5.0% 4.5% 4.0% 3.5% Current price Proportion of trades 3.0% 2.5% 2.0% 1.5% 1.0% 0.5% 0.0% -0.46-0.43-0.41-0.38-0.36-0.33-0.31-0.28-0.26-0.23-0.21-0.18-0.16-0.13-0.11-0.08-0.06-0.03-0.01 0.02 0.05 0.07 0.10 0.12 0.15 0.17 0.20 0.22 0.25 0.27 0.30 0.32 0.35 0.37 0.40 0.42 0.45 Source: Bloomberg Trading price relative to current price Options can also be used to achieve better levels when executing simple directional trades. It is quite common to hear a manager say something like, If the market reaches x, then I m going to sell it. If that is really what the manager will do, then instead of waiting for the chance that the market does reach the target, they will produce a better result by selling short dated call options on the market with the strike at the target level. If the market does reach the target, then they will effectively sell it as the calls are exercised. If the market doesn t reach the target, then they will make money from the option premium. Options often trade at higher volatility than the actual volatility of the market, so the options provide a better return over time. 13
Example option position The chart below illustrates one way of using options to implement a view to be long duration at a level of 2.00%. As shown, through selling a short dated European put option at this level, a fund would be able to earn the premium (0.10%) should yields remain low, but also be long duration at the desired level should yields end up higher than 2.00%. In this example, yields have remained low and the option would have expired, however the fund would have realised the premium in addition to expressing their duration view over the period. Chart 6: Expressing a long duration view via options 2.20 Break even level Bond future yield (%) 2.10 2.00 1.90 1.80 Put option sold, long position trigger at 2.00% 1.70 10 Mar 24 Mar 07 Apr Premium earnt Option expires, premium realised 3 year bond future yield Sold 1x Put Breakeven level For interest, in the chart below, the future distribution of trading levels is overlayed on the past distribution of trading levels from the chart above. The distribution of trading levels was similar to what would be expected. Over time, using the entire distribution of levels will add value. The value added from these activities will typically be attributed as part of the duration excess returns. Chart 7: Distribution of 3 year Australian bond futures trades (4 months to May 2016 vs 4 months after May) 8.0% 7.0% 6.0% 4 months to May 4 months after May Proportion of trades 5.0% 4.0% 3.0% 2.0% 1.0% Current price 0.0% -0.46-0.43-0.41-0.38-0.36-0.33-0.31-0.28-0.26-0.23-0.21-0.18-0.16-0.13-0.11-0.08-0.06-0.03-0.01 0.02 0.05 0.07 0.10 0.12 0.15 0.17 0.20 0.22 0.25 0.27 0.30 0.32 0.35 0.37 0.40 0.42 0.45 Source: Bloomberg Trading price relative to current price 14
Conclusions Traditional fixed income portfolio managers can add value, and they typically do, when the targets are conservative. In enhanced funds, they use many small, diversifying, and often sophisticated methods. However, when alpha targets rise, this sophistication often gives way to a simple view on the direction of interest rates: up or down. It need not be like this. There are plenty of opportunities for many small, diversified and sophisticated ways to add value through directional and relative value trades and plenty more ways to add value through an array of directional strategies, as we have demonstrated. This allows active managers to achieve higher alpha than typically available in enhanced funds without giving up on consistency. The manager will need to be appropriately skilled in dealing with these and other approaches to ensure that each is managed properly and the overall portfolio structure containing these strategies is efficient. 15
For more information call Macquarie Investment Management on 1800 814 523, email mim@macquarie.com or visit the website at macquarie.com/investmentmanagement BM-1095 0417