+1 Added to my documents.
Please be aware your selection is temporary depending on your cookies policy.
Remove this selection here

US HY default rates: trends, projections and perspectives


The essential

The current low default rate regime of US HY issuers is the longest in recent decades and has also survived the latest commodity-driven mini cycle: the latter is close to its end and short-term expectations point to a further fall of the DR to around 3% in the coming quarters.

In this piece, we focus on both a one-year horizon and a longer perspective, analysing the most relevant top down and bottom up factors leading US HY extreme credit events. As rising bond yields and monetary policy changes are increasingly on credit investors’ radars, we have analysed the link between real and nominal bond yield levels and the yield curve slope on one side; and speculative grade default rates, on the other, in order to address the question about their potential impact on the current default cycle.

Download Thematic Paper


27 September 2017


27 Septembre 2017


On-going trends

A steep dive in US HY default rates recorded in H1 2017

H1 2017 saw default rates of US speculative grade bonds turn sharply south: the trend was mostly expected, as all major indicators leading bankruptcy rates of US HY companies were pointing to a peak at the very beginning of the year and to a subsequent, rapid fall. As recent Moody’s data confirm, the fall has actually accelerated in the latest months of 2017. January saw a 5.9% DR for US speculative grade companies, the peak of which may be referred to as the current default “mini-cycle”. Then, the rate fell to 4.7% in March and touched 3.8% in June: the latest rating agency forecasts point to around 3% for December this year, which is very much in line with our own regressed forecasts, based on indicators leading defaults by at least one year. A gentler drop should therefore take place in the coming months, as the steep fall of the trailing 12-month default rate by June took place because of the substitution of a very bad H1 2016 with the much better numbers recorded in H1 this year. In terms of defaulted debt, in fact, in
the first six months of 2017, US HY recorded less than one third of the amount that occurred in H1 2016, while the second part of this year is expected to diverge less from the same period last year.



The recent commodity-driven mini cycle is close to its end

Both upward and downward trends of the current “mini-cycle” were not recession and/or financial crisis-driven but, instead, almost entirely commodity-driven: if energy and material sector companies are excluded from calculations for the last two years, default rates actually remained quite stable and close to the low levels of previous years. Chart 1 shows that the recovery of the energy sector in terms of decreasing number of bankruptcies should proceed in the coming quarters, lagging by around one year the massive drop in its distress ratio. The distress ratio the percentage of bonds trading above a 1,000 b.p. spread over treasuries) of the entire US speculative grade universe is currently close to the lows of 2014, therefore pointing to a 3% threshold in six to nine months’ time. A look at defaults by rating categories (chart 2) is encouraging, too. High quality speculative grade bonds, namely the BB-rated universe, have moved down to a zero DR since January this year, while B-rated bonds declined to the same level over the subsequent months. In a nutshell, therefore, default rates are currently concentrated in the lowest rating
category only, namely the CCC-rated names.



Short term projections

At the end of the day, from a top-down approach, default cycles are very much a macro growth story and a story of financial conditions.

Macro growth. The link between macro growth and speculative grade default rates is intuitive and sees peak years and default trends very much dependent on changes in the GDP trend. Interestingly, however, despite GDP growth that has slowed down significantly since 1990, the default cycles proved to be less and less painful for US speculative grade companies, confirming something which is not so intuitive, namely the fact that it is not the level of macro growth that affects defaults. Historical evidence, instead, shows that it is not the level of growth but the change in GDP growth with respect to its long-term trend that is very much the real driver of the default cycle. In this respect, our forecasts point to an encouraging stable outlook for US growth over the coming quarters, in the 2% area: despite the prolonged growth cycle and the low level of real growth, the risk of a downturn remains quite subdued, considering most indications coming from macro leading indicators.

Financial conditions are the other major driver of default rates, together with macro growth. This is because low-rated companies are highly dependent on external funding and vulnerable to sudden changes in liquidity conditions and investors’ risk aversion. As the bond market has become a major funding channel for US speculative grade companies in comparison to bank loans, the role of financial conditions in the bond market has also become increasingly powerful, as a driver of defaults.



Actually, the most powerful inputs of our forecasting models are represented by both the trends in financial conditions for bank loans and speculative grade corporate bonds. The availability of bank loans is captured by the survey conducted and published by the Fed on bank lending standards on a quarterly basis, while the openness of the bond market to fund speculative grade companies may be measured by the distress ratio, or the percentage of HY bonds trading above a 1,000 b.p. spread over Treasury bonds.



Interestingly, both factors lead default rates by around four quarters on average: intuitively it makes sense, as the unavailability of credit “drought” of liquidity and subsequently contribute to an extreme credit event. As we already mentioned, both factors are currently pointing to default rates declining to around 3% in one year’s time (see chart 3).

Let’s now move on to bottom up drivers

As top down factors point to a further fall of the DR and commodity sectors’ woes recede, the question is now about which sectors may move to the frontline of defaulted issuers and lead the trend in the next months and quarters. At the moment, excluding commodities, the retail sector is showing some stress, although with just a few defaults recorded last month and a distress ratio at 27%, which is currently the highest among US major segments. According to the latest figures published by Fitch, however, the US retail sector is currently suffering from a still very limited DR, namely 1.8%. Moody’s is forecasting a 4.5% DR in one year’s time for retailers, while the Media sector has been recently revised up to 7% in 12 months’ time: in a nutshell, this is a very limited rise for the two sectors, respectively representing 3.7% and 4.7% of US HY bond market debt, a lower weight than the 14% still represented by energy and the 5% by metals & mining. Widening the analysis to the other sectors currently recording the highest distress ratios, the picture does not look so worrying: together with media, capital goods and telecoms are showing distress ratios that are falling and pointing to lower bankruptcies in one year’s time


A look at recent trends in debt supply sees refinancing recovering ground on M&A and spending items among the purposes of new bonds: this is an encouraging sign, in light of US companies’ limited refinancing needs in the next two years. Also in this respect, no major concentration of maturing debt seems to be in the pipeline over the coming quarters. In terms of credit quality, finally, CCC-rated debt remained quite contained in this cycle within the US HY benchmark: after the Lehman crisis, in fact, the lowest-rated debt never exceeded the 20% threshold and actually moved down on the back of the commodity crisis to its current 15% level. The last two default rate spikes in the US saw CCC-rated debt reach one quarter and one third of the overall market, respectively (see chart 4).

A longer perspective

Long live this cycle, 13 years old! The current low default regime also survived the recent commodity-driven mini cycle

Apparently in a quite paradoxical divergence between macro and financial trends, the current cycle of HY default rates in the aftermath of the Great Financial Crisis continues to be the most benign since 1990. Chart 5 shows annual default rates of US B-rated names over more than the last three decades. Both ten-year averages and the worst yearly defaults followed downward trends. Interestingly, ten-year default averages fell from 8.5% in ‘84-‘93 to 4.2% in ‘94-‘03 and finally to just 1.5% in the thirteen years ending in 2016. The worst yearly defaults also fell from 15% (1990) to 9.5% (2001) and finally reached “just” 7.4% in 2009. What is most striking about the performance of the latest regime, however, is that default rates surpassed the 5.2% long-term average only once, while they remained between 0% and 1% in nine of those years. Defaults of Ba-rated issuers also show a very similar pattern, while defaults of CCC names suffered more than other rating categories from the last commodity-driven mini cycle.

We all know that the very good performance of the six years following the peak of the crisis, namely the period between 2010 and 2015, had certainly to do with an unusual phase of abundant liquidity and search for yield, ultimately supporting the demand for speculative grade bonds and therefore keeping default rates from rising. However, despite the recent rise on the back of the 2014 commodity crisis, default rates also remained quite low by historical standards in recent years, especially among high and medium rated names.

In the previous section focused on top down factors leading defaults by one year, we already mentioned the effective role played by BLS indications and distressed ratios. However, financial conditions are not only measured in terms of access to credit and volumes of funding available to companies in both bank and bond market channels: they also depend on the cost at which funding is made available to corporates. The level of rates with respect to real and nominal growth, furthermore, seems to determine even more the nature of the default cycle.


In this respect, the experience of the ‘80s, the ‘90s and the early ‘00s had all shown real rates rising to the 2%/3% area before the last three default cycles peaked. Moving from real to nominal yields, the link between defaults on one side and yields paid by HY bonds over almost the last 40 years seems to point to
quite a positive correlation. Chart 7 plots together the yields and default rates of US B-rated debt (B-rated debt represented and still represents the average credit quality of the overall market benchmark).

The two variables moved quite in line, underlying that in an ultra-low yield environment default rates steadily fell below long-term averages and also failed to regain ground in the latest mini cycle. As we have already mentioned in a previous paragraph, this link between yields and defaults over the last 40 years or so is very much similar to the one between real growth and defaults. In a nutshell, therefore, a low macro and micro growth, low real and nominal yield environment look quite supportive in keeping defaults of speculative grade bonds from rising.

The key question to address now is whether a rising bond yield environment threatens the return of the default cycle for speculative grade bonds.

As the Fed proceeds further down its path of normalising monetary policy, funding costs could therefore represent a key variable, likely to take centre stage on credit investors’ radars. Specifically, market participants are increasingly asking themselves about the potential threat to corporate bonds arising from the future increase in yields and cost of debt. This potential threat from higher yields is twofold: on one side, rising yields could reduce the relative attractiveness of corporate bond valuations vs. underlying risk-free assets (this is the topic of another credit focus) and also support a rotation in favour of equities. On the other side, higher bond yields may increase fundamental risks, leading to a renewed default cycle.

To address the question, the chart 6 shows to what extent a rise in real rates preceded the default rate spikes of the previous cycles: at the same time the chart also underlines the fact that the current level of US real rates is still a long way from those highs of the past decades and the new macro-financial environment in place after the GFC (huge debt, ultra-easy monetary stimulus, low growth, low inflation regime) is unlikely to experience the same levels of real yields again in the medium term. Simply put, we would need the cost of funding to rise so much to become structurally higher than growth/inflation in order to get back to pre-crisis default rate standards. Despite normalising its monetary policy, the Fed is unlikely to conduct a “disruptive” spike in bond yields which would be needed to spur a renewed default cycle: the growth potential of the US economy is lower and this is going to keep yields from rising to pre-crisis levels.

In order to fully address the potential effect from financial conditions, we also investigated the link between defaults and the yield curve slope. The latter tended to lead bank lending standards and distress ratios by between four and six quarters, both of which in turn lead defaults by around four quarters. As chart 8 shows, the past link has become weaker in recent years, and the slope is not yet in areas which in the past triggered a tightening of credit conditions. In fact, a steep curve forces investors to move down the risk structure and out of the term structure, therefore increasing risk appetite: on the contrary, a flatter yield curve reduces opportunity costs of risk averse investors, supporting a progressive switch out of spread products back to risk-free bonds. However, in an ultra-low yield regime, the opportunity costs for risk aversion look higher, as financial repression prevents the curve and the absolute yield levels from moving as they did in the past.


  • Despite most of the fall from their recent peaks having already occurred in H1 2017, the default rate of US speculative grade companies is likely to continue on its current downward path over the next six to 12 months.
  • In light of current distress ratio levels, no major sectors look to be forging ahead to fully replace the energy sector in leading a similar new cycle in the coming quarters. Retail became an increasing contributor, but its limited weight should reduce its impact on the overall universe. Moreover, to some extent, the last two years probably added some incremental longevity to this cycle by partially wiping off some of the forming excesses.
  • Despite being in place for 13 years, the current low default regime is also likely to persist beyond the 12-month horizon: on one side, macro growth is expected to remain around 2% in 2018, while on the other side financial conditions do not seem to pose major threats to a stable picture.
  • Despite the Fed moving down its policy normalisation path, a low growth/ low inflation scenario should keep real and nominal yields from rising back to levels which proved to be dangerous in the past.
  • The yield curve slope also looks flatter than a few quarters ago, but it remains far from flat or the negative numbers which, in the past, helped trigger risk aversion.















The current low default regime
has also survived the recent
commodity-driven mini cycle.

























Since 2004, B-rated default rates have
surpassed the 5.2% long term
average only once, while they
remained between 0% and 1% in nine
of those years.





















CCC-rated debt remained
quite contained in this cycle
within the US HY benchmark
















































One of the elements to follow at the September FOMC will be the evolution of the "dots"













BERTONCINI Sergio , Head of Rates & FX Research
Send by e-mail
US HY default rates: trends, projections and perspectives
Was this article helpful?YES
Thank you for your participation.
0 user(s) have answered Yes.