Credit: Thinking beyond the benchmark
D William Kohli discusses the need for a broader opportunity set in a rising rate environment
For several decades, the Barclays Capital US Aggregate Bond index (known as the Lehman US Aggregate Bond index until November 2008) has been a central reference point for bond investors - a benchmark with widespread acceptance, comparable to the S&P 500 in the equity world.
Barclays’s global counterpart, the Barclays Capital Global Aggregate Bond index, has more recently achieved a similar stature on the global stage: the fixed-income version of the MSCI EAFE. Between passive exchange-traded funds and index funds, actively managed mutual funds and institutional mandates, many trillions of dollars are invested in strategies pegged to these two benchmarks, and too often, this fact alone results in a significant misallocation of risk for investors.
The core belief at the foundation of any relative-return strategy is that risk stems from a portfolio’s deviation from the benchmark. It is the portfolio manager’s decisions about what sectors or industries to overweight and underweight, which securities to own and which to avoid, that drive all meaningful metrics of risk and return, from standard deviation to alpha to information ratio. Because the bond markets, as proxied by the Barclays Agg indices, are generally thought to be efficient, conventional wisdom has it that outperforming the benchmark can only be achieved by assuming a greater degree of risk — risk reflected by a portfolio’s deviation from the allocations of the benchmark.
This belief is flawed for several reasons. There are a number of risks inherent in individual fixed-income securities, from term-structure to credit risk to price or liquidity risk. The potential returns associated with those risks, accordingly, fluctuate both over time and depending on the characteristics of any given security. In other words, not all risks are — or should be — considered equal. But at the index level, many of those risks that characterise the broad bond market are under-represented or entirely absent. The result for any strategy that uses the Barclays Agg as its starting point, we believe, is a significant over-allocation to those sources of risk and sectors of the market that offer the least potential for returns in today’s market.
The key problem with using the Barclays Global Agg to guide risk allocation decisions lies in the construction of the benchmark. Because the Global Agg is designed to reflect the broad bond markets as a whole, the largest issuers, naturally, represent the lion’s share of the benchmark. But as the equity markets have shown in the 2000s, this kind of blind bias towards market capitalisation can have undesirable consequences.
In an era of soaring government deficits across developed nations, the magnitude of the allocations to government debt in the Barclays Global Agg is striking: At the close of 2010, more than half the index was invested in Treasury bonds, and almost 85% of the index was allocated to government-backed securities.
An unprecedented period for bond market intervention is coming to a close. Less than a year ago, the mantra - especially in the US - for bond investors was to ‘own what the government was buying’. Today, the opportunity to front-run central banks is largely gone as global bond markets approach the end of a period of government intervention that was historic in size and scope. As the dust has settled, we now have an unprecedented fixed-income landscape:
• Between November 2008 and March 2010, the US Federal Reserve purchased $1.25trn of agency mortgage-backed securities (MBS), representing about 25% of the outstanding market.
• In May 2009, the European Central Bank began a €60bn purchase programme aimed at euro-denominated covered bonds.
• Between October 2010 and June 2011, the Fed implemented a second round of quantitative easing, purchasing $600bn in US Treasuries, adding to the approximately $300bn in US Treasuries already on its books.
• The US federal funds rate and ECB policy rate currently hover near record low levels.
From a public policy perspective, it is hard to argue that developed economies’ central banks could have taken any other path, given the dire condition of the credit markets at the time. However, any such dramatic, one-sided market activity produces massive distortion, and we see that distortion today in the relative overvaluations of Barclays Agg sectors that were targets of central bank purchases. Yields on government securities, agency debt and agency MBS are at historically low levels, and spreads in a number of sectors are tight by historical standards. Prices on such securities have been propped up by massive artificial demand, leaving little upside potential, but great downside potential as central banks’ easy money policies are recalibrated to expectations of economic growth and inflationary pressures. In short, as the era of easy money comes to an end, the risk of rising interest rates across developed markets is substantial.
The risks most likely to offer compensation are under-represented in the benchmarks. Given what we know about the forces at work in global bond markets, investors would likely seek to limit the term-structure (or interest-rate) risk in their portfolios. But again, because of the construction and nature of the benchmarks, the overwhelming majority of the risk exposure they assume is to term structure and, in the case of the Global Agg, the highly correlated foreign exchange risk.
If investors are willing to accept the premise that a portfolio that bears little resemblance to the Agg may actually carry less risk, a number of new opportunities become available. More than $4trn (€2.8trn) worth of securities is not included in the Barclays Global Agg, which offers more than twice the average yield of the benchmark and an average option-adjusted spread nearly eight times higher. Moreover, the source of the risk inherent in those sectors is far more diverse and much less contingent on declining interest rates to fuel returns.
US non-agency residential mortgage-backed securities (RMBS) represent a prime example of one of the opportunities available to active managers willing to look beyond the benchmark. As securities that lack any government support, US non-agency RMBS returns are driven by mortgage credit risk, prepayment risk and term-structure risk — in that order - making them significantly less sensitive to changing interest rates than the components of the Aggregate indices.
Since the credit crisis, it has become clear that mortgage underwriting standards were often shoddy. Many investors and rating agencies did not understand the great risk embedded in housing prices, as losses have torn through the MBS capital structure, reaching even the highest tranches.
Today, prices of US non-agency RMBS have been ‘marked down’ severely, often exceeding the extreme declines in the value of the underlying collateral. As is frequently the case, the market has swung from extreme overvaluation initially to the undervalued condition we have today. Despite the fact that top-tier MBS are first to be paid, many continue to trade at 60 or 70 cents on the dollar. Whilst those securities may never return to par value, we believe there is significant opportunity for capital appreciation, even if housing prices continue to decline and defaults increase.
There are a number of other areas we believe offer the potential for diversifying away from term-structure risk and also offer the opportunity for further spread tightening relative to US Treasuries. Importantly, these sectors all share a common characteristic: They represent tiny fractions of the benchmark Aggregate indices.
Active management is the key to generating attractive returns in a rising rate environment. Although we believe that the Barclays Aggregate indices represent a poor allocation of risks given today’s historically low interest rates and the massive government intervention in the bond markets, that is not to say that all term-structure risk is undesirable. There are ample opportunities for active managers to use discrepancies in interest rates and global growth rates to their advantage. Over the past five years, for example, economic growth in select emerging markets — including India, Taiwan and South Korea — has significantly eclipsed that of bellwether developed markets including Japan, France and the US.
This divergent global economic growth is ready-made for the carry trade or establishing relative value positions. But again, these opportunity sets exist only outside the narrow context of the Aggregate indices, which have no significant exposure to below-investment grade or emerging market securities.
Putnam has the resources necessary to pursue a truly active strategy. Whilst the Barclays Aggregate indices are, of course, likely to remain key touchstones in the bond market, we believe investors need to understand the risks inherent in a benchmark-centric investment strategy, particularly in light of the historic events sparked by the 2008-09 financial crisis.
Putnam favours an approach to fixed-income investing that targets opportunities across the bond markets, both inside and outside the index, whilst allocating assets towards those risks most likely to offer attractive returns. In today’s investment environment, with interest rates kept artificially low in a number of developed markets, we believe the potential returns for assuming term-structure risk within the constraints of the government-heavy Barclays Aggregate indices are not commensurate with the significant downside risks entailed.
Investing outside of the Barclays Agg indices requires the kind of substantial expertise, diligence, and resources that Putnam has for decades devoted to this type of active management investment approach.
D William Kohli is co-head of fixed income for Putnam Investments and team leader for portfolio construction