- Actions speak louder than words. The most prudent and objective approach to understanding a manager’s investment style is to perform a rigorous statistical analysis of their long-term historical track record. Such an approach is extremely revealing. In contrast, market chatter is often misleading.
- Western Asset’s broad market core product has outperformed the Bloomberg Barclays U.S. Aggregate Bond Index by an average of 140 bps gross of fees per year over the 42-year history of the index.
- This paper thoroughly examines our historical track record using a number of statistical processes. We found no evidence of any systematic bias. Our record is one of outperformance through both risk-on and risk-off markets.
- Despite misconceptions that we are “just a credit shop” or exclusively a “risk-on manager,” this product also outperformed the index (gross of fees) in 139 of the 149 rolling five-year periods—or 93% of the time.
- We acknowledge and own our poor performance during the financial crisis of 2007–2008, but we rebounded quickly and completely—recovering all lost ground in two years—and learned from the experience.
As a senior leader of the investment strategy process at Western Asset for over 27 years, I’m very gratified with the returns we’ve been able to generate for our clients. While there are no guarantees, at Western Asset we feel very strongly that our value-driven investment process combined with our disciplined pursuit of diversified strategies should continue to produce the results our clients have come to expect.
Exhibit 1 displays Western Asset’s Core Full Capabilities track record since the 1976 inception of the Bloomberg Barclays Aggregate Bond Index (the Aggregate). Western Asset’s broad core product has outperformed the Aggregate by an average of 140 basis points (bps) gross of fees per year over the 42-year history of the index. Despite the enormity and variety of market upheavals and environments, our results are significantly positive for all cumulative periods shown. Most particularly, even the last 10-year period, which includes the 2008 crisis when we struggled mightily, shows a compound average outperformance of 170 bps.
At Western Asset, our focus on highly diversified portfolios means we should succeed if we are right more often than we are wrong and if we make more money when we are right than we lose when we are wrong. If we successfully use our extensive global platform to build exceptionally diversified portfolios, we should minimize risk as well. Whether the market is risk-on or risk-off, whether interest rates are going up or down, our performance should be a function of our idea generation and portfolio structuring effectiveness. Indeed, over the last 42 years, this has been true of our record.
Of course every firm has a narrative to support why it expects to perform well. Thoughtful investors, in turn, utilize statistical analysis of investment track records to understand and verify whether the narrative conforms to the historical facts. What has always struck me, though, is just how powerful the narrative about a firm or its process can be—even when it can be confirmed, or not, by simple fact-checking. I’m reminded of the book “Moneyball,” based on the true story of how using basic baseball statistical analysis, which had been widely understood for over 20 years, was implemented with such success that it revolutionized the baseball industry. My main point of interest from this book has always been just how difficult it was for the protagonist to implement change—even though it was obviously needed and ultimately very beneficial. The statistical arithmetic was unassailable, yet virtually every baseball insider defended the status quo method that relied heavily on intuition.
So it is with investment firms. Impressions of them formed by outsiders become sacrosanct, sometimes regardless of whether or not they square with the facts. Investors, commentators and consultants often grasp for the readily available overview of a manager’s performance based on common industry hearsay: “This firm is good at credit, or macro, or mortgages, or emerging,” or “This other firm can’t manage risk, doesn’t know anything about interest rates, or has terrible analytics.” So much of that industry talk is just false, persisting despite clear evidence to the contrary.
At various times throughout my tenure, Western Asset has been charged by some critics with “hugging” the index or getting too far away from the index, taking too much interest-rate risk, or not taking enough interest-rate risk, and of course, taking too much spread-sector risk. All this comes with the territory of money management. Despite enviable performance and impressive risk-adjusted returns, we have had periods of underperformance and have made our share of investment misjudgments. These ups and downs should not be dependent on any particular market environment, however. Results should merely reflect the weighted returns of our diversified strategies. Were we right more than we were wrong?
With market chatter sometimes developing a life of its own, we have often been confronted with the perception that Western Asset is overly reliant on credit, with a risk-on bias. Now, there is nothing inherently wrong with running such a program. Indeed, many managers use this technique successfully. This approach, though, would have a strong tendency to underperform in credit-challenged or risk-off periods. This is simply not the case at Western Asset because we do not have a systematic risk-on bias.
Historically our performance during risk-off periods has been strong, with one notable exception—the financial crisis. I suspect that the false narrative regarding Western Asset has its roots in our challenged performance during this period. Yet we rebounded quickly and completely, recovering all lost ground in the broad market core product in two years. In the period since, as with the periods prior, our performance was not only substantially positive, it also bears no statistical evidence to support the label of a substantial risk-on bias. This is the point we really have to reinforce. Fortunately, just as in baseball, the statistical tests and frameworks for analyzing track records have been around for decades. In our case, the perception based on the false narrative that Western Asset is “just a credit shop,” or overly reliant on spread product, or always “risk-on” can be easily debunked by statistics.
This paper runs through basic performance tests that shed light on a manager’s strengths and weaknesses, using Western Asset’s long-term Core Full Capabilities1 performance track record. The Core Full Capabilities performance record represents the long-term investment results for Western Asset’s US-based broad market portfolios, reflecting the evolution of the Core (and Core Plus) investing style as it expanded its scope of permissible instruments (such as derivatives, below-investment-grade bonds and non-US bonds). The record reflects approximately the last 25 years of Core Plus composite performance—a more detailed explanation, along with the standard Core Plus composite information, is in footnote 1 and in the Appendix. I believe that focusing on performance “gross of fees” is the best way to demonstrate our investment ability. However, to be complete, we also show performance “net of fees” deducting the highest current standard fee rate. The results are informative, as they provide a compelling rebuttal of the most common misconceptions we sometimes hear. Obviously, we have a clear self-interest in championing our results, but the statistical tests are straightforward, and the results speak for themselves.
Test of Time
The simplest gauge of all is the test of time. Historical performance shows how a firm has navigated the vicissitudes of global economic and market forces. The longer the track record, the more obstacles a manager has had to endure and/or overcome. Delivering 140 bps of annualized excess return, gross of fees, for over four decades is a record hard to match in our industry. When you consider the enormous changes to the investment landscape that have occurred throughout this period, such substantial outperformance is an indication of a strong resilient investment program.
Test for Risk
The standard test for risk/reward is displayed in Exhibit 2 along with the usual accompanying metrics. The standard deviation of returns displays how much risk a manager takes in order to achieve better returns. This is crucial for fixed-income investors because bonds are the defensive asset class. Managers are charged with generating excess returns while approximating benchmark risk. The information ratio measures incremental returns from a strategy against incremental risk.
Over the last 42 years, Western Asset’s 140 bps of average excess gross returns came with an additional 253 bps per year of tracking error against the Aggregate, for an information ratio of 0.56. Notice also that the tracking error figure reported in Exhibit 2 somewhat overstates the incremental risk from Western Asset’s Core Full Capabilities strategy versus the Aggregate, as the standard deviation (volatility) is only 19 bps per year higher than that of the Aggregate.
This test also benefits from longer time periods. The sources of risks and returns available in the market vary widely across business and market cycles. The passage of years and decades means strategies well-suited for one time period that aren’t changed in a timely fashion will introduce a meaningful increase in risk. Significant periods of highly volatile performance or substantial underperformance work to reduce both average excess returns and the information ratio. Over a long period of time, information ratios of 0.5 or higher are very difficult to achieve and clearly signal a strong focus on risk-adjusted returns. Keep in mind, too, that our information ratio stands despite the negative performance during the crisis.
The Historical Consistency Test
Rolling five-year returns can be a good gauge for consistency. A manager may have had success in some time periods but not in others. More to the point, sometimes a good strategy has to be held through tough times before it bears fruit, while other strategies must be cut loose before losses persist. Looking at five-year returns helps determine how proficiently a manager has stuck with slowly developing strategies while getting out of the ne’er-do-wells.
This test is particularly useful in identifying overreliance on specific skill sets. This will show up as outperformance when those skills are favorable and vice versa. For instance, a manager who places a heavy reliance on credit would plausibly outperform the benchmark during the upturn in the business cycle but underperform in the downturn. There is nothing inherently wrong in this either. Many skilled credit managers exhibit just this pattern.
If such credit managers make more money during the longer upturn cycle than they lose in the occasional downturn, they are still delivering meaningful value over the entire cycle. They will, however, show negative periods or “clusters” of underperformance during the business downturns. Whenever a recession hits, you will see negative returns that will drag down rolling five-year numbers. So seeing positive long-term results with negative clusters during risk-off periods is strongly suggestive of a heavy credit focus.
Exhibit 3 displays the annual benchmark and our performance as well as the rolling five-year excess return versus the index for the Western Asset’s Core Full Capabilities track record over the last 42 years. There were 149 rolling five-year period results, and Western Asset showed an outperformance (positive result) on a gross basis 139 times or in 93% of them3.
Notice that the only negative cluster is focused around the financial crisis years of 2007–2008, and the rolling five-year excess return was back into positive territory by the end of 2010 (thus fully offsetting 2008 underperformance).
Again, in all other recessions and risk-off periods, you see no negative clusters. This strongly supports the absence of a risk-on bias. For a diversified manager, this is not surprising. For a manager heavily reliant or biased to be overweight credit and other spread products, the scarcity of negative clusters would be virtually impossible.
Let’s look at specific risk-off episodes of the last 42 years as either risk-on or risk-off, based on whether excess returns of credit and emerging market (EM) spread products were negative or positive4. For the 42-year record of the Aggregate, there have been 15 risk-off years.
Excluding the crisis years, Western Asset’s Core Full Capabilities track record was down between 100 bps and 200 bps twice, down between 50 bps and 100 bps once, down less than 50 bps twice, up less than 50 bps four times, up between 50 bps and 100 bps once, and up between 100 bps and 200 bps thrice. In other words, we showed more ups (eight) than downs (five), better performance in the ups than in the downs, and no undue stress. This echoes the message from the five-year rolling excess returns.
Scrutinizing the Narratives
Let’s examine the “Western Asset is a risk-on manager” narrative. The Aggregate contains “official” excess return data starting in August 1988. Exhibit 4 displays the complete description of excess returns for investment-grade credit, high-yield credit and EM for every risk-off year in the last 20 years, excluding the financial crisis of 2007–2008.
In 1998, there was the Asian contagion as well as the Russia and LTCM defaults. Risk was off. EM underperformed Treasuries by more than 20%—or 2,000 bps—and high-yield underperformed by more than 8%. If the narrative were true that Western Asset is highly reliant on credit, we might have been expected to have an awful year. Instead, we outperformed the Aggregate by 31 bps.
In 2000, the NASDAQ collapsed, a recession started, high-yield fell by almost 19% and investment-grade was down over 4.5% versus Treasuries. A credit shop must have suffered mightily, yes? Not Western Asset; we were up 38 bps versus the index.
2001 was a continuation of the recession, with the 9/11 tragedy to boot. This was a more modest risk-off period, with investment-grade and high-yield both down less than 3%. Western Asset beat the index by 147 bps.
2002 saw the WorldCom default and the demise of Arthur Andersen. Corporate bonds hit the skids, with high-yield down over 13%. Was this a crisis for Western Asset? No, we had a small underperformance of -37 bps.
2011 was a risk-off year following the European peripheral crisis and there were fears that it would spread to the US banking sector. At Western Asset, we had another modest minus year of -52 bps.
2014 was a modest risk-off year with all three sectors showing moderate negative excess returns. Western Asset had an excellent year, outperforming the Aggregate by 173 bps. In fact, we actually won Morningstar’s Manager of the Year award6—in a risk-off year.
Last, 2015 brought the commodity price collapse and fears of a global slowdown led by China. Global and US credit, particularly US high-yield, performed very poorly. Once again we posted outperformance, this time of 64 bps.
Taken together, these seven years of very challenging risk-off environments would have bedeviled a highly concentrated credit program. Western Asset’s Core Full Capabilities track record had no meaningful underperformance and saw outperformance in five of those years, with a cumulative gross outperformance of 364 bps7. So much for the idea that Western Asset is exclusively a risk-on manager.
Let’s Put the Shoe on the Other Foot
Is it better to have good long-term performance hindered by a single adverse episode, or modest performance boosted by a single positive event?
At Western Asset, we have been fortunate to post positive performance for our clients across all long-term time periods for our broad market core product. Detractors can rightly point to the financial crisis as a difficult time for us. But all our long-term statistics are inclusive of the crisis. Yes, our numbers have been diminished by the crisis, but imagine the opposite.
A manager who by foresight or simple caution avoided the crisis has long-term numbers meaningfully elevated by this single event. But if, in checking the statistics, it were the case that most of their long-term performance arose from the 2008 crisis, then the opposite narrative is suggested. If Western Asset, based on historical evidence, should be regarded with caution by those fearing a crisis, such a manager should be valued only by those absolutely counting on another crisis.
Woulda Coulda Shoulda
We finish up with a flight of fancy. The crucible of change is always upon us, and the resilience to bounce back from adversity is crucial. So the following is a somewhat whimsical statistical review, but bear with me.
In our business, everyone has to examine the question when confronted by underperformance: Is this a good team that suffered a bad outing or a bad team we need to let go? The statistics below answer this question decisively. We are an excellent team that had a bad outing.
We reran all the statistics, removing the years 2007–2010. Western Asset had bad performance in 2007 and 2008, followed by a spectacular rebound in 2009 and 2010, leaving investors roughly flat for the period. Let’s look at the 38 years of our experience excluding the 2007–2010 fall and rebound.
For those 38 years, our annualized excess return was 141 bps, with an unheard-of information ratio of 0.98. Five-year rolling excess returns were positive in 131 of 133 periods, over 98% of the time8.
Markets are incredibly difficult. The competition in our industry is fierce. Any study of capital market theory reveals just how brutally challenging it is to outperform the market even for a short while. To do so in a very meaningful way over 42 years is an accomplishment we are exceptionally proud of.
More importantly, fixed-income is a defensive asset class. Investors want strong risk control, emphasizing mitigation of drawdown versus benchmarks. This is a daunting task. The goal is to beat the market, beat your competitors, limit your downside risk and develop significantly positive risk-adjusted returns.
How have we managed to accomplish this task over virtually the entirety of the existence of the aggregate index? All investment managers have their ups and downs. But if you have a strong investment philosophy and process—and are sufficiently resilient and introspective to make changes as markets evolve—your chances of favorable long-term results are good. Indeed, we believe this is what the statistical evidence clearly demonstrates.
- The Core Full Capabilities performance presented is supplemental information generated from actual portfolio performance that represents the evolution of Western Asset’s investment skills over the history of the Firm’s core fixed-income investment style. From inception through 31 December 1988, performance is represented by accounts that participated in the US Core Investment Grade No Futures & Options Composite. Beginning 1 January 1989 and through 31 December 1990, our capabilities expanded to include futures and options, and therefore this performance consists of accounts that participated in the US Core Investment Grade Futures & Options Composite. From 1 January 1991 through 31 January 1993, performance is represented by accounts participating in the US Core Below Investment Grade Futures & Options Composite to reflect our expanded capabilities to include below investment-grade securities. Beginning 1 February 1993 to current, we started utilizing opportunistic investment in non-US$ securities, and therefore this performance consists of accounts participating in the US Core Full Below Investment Grade Futures & Options Composite.
As of 31 Dec 17, returns for periods greater than one year are annualized. The Core Full Capabilities performance provided is supplemental to the US Core Full BIG F&O Composite. Please see performance disclosure in the appendix. Returns illustrated should not be considered an indicator or guarantee of future results and should not constitute the sole basis for an investment decision.
- Net of fees (01 Jan 76 - 31 Dec 17): Annualized Return 8.53%; Standard Deviation 3.32%; Tracking Error 2.53%; Information Ratio 0.43.
- On a net of fees basis, Western Asset showed an outperformance 133 times or in 89% of them.
- Given the dynamic market structure—high-yield data were not completely reliable in the 70’s, EM data unavailable until the 1990s—the scoring was done in terms of available sectors.
- Morningstar Awards 2014 ©. Morningstar, Inc. All Rights Reserved. Awarded to Ken Leech, Carl Eichstaedt, and Mark Lindbloom, Western Asset Core Bond and Western Asset Core Plus Bond for 2014 Fixed-Income Fund Manager of the Year, US. The funds are offered through Legg Mason Investor Services, LLC—www.leggmason.com and are not available to non-US investors.
- Net of fees cumulative outperformance was 140 bps.
- Net of fees annualized excess return was 1.12%. Using net of fees, the five-year rolling excess returns were positive 129 of 133 periods, over 97% of the time.