There is more to come for high yield

In 2006, a host of issuers, whose credit ratings effectively barred them from accessing the unsecured debt markets, turned to bank loans, which at the time, were priced about 100-150 bps below bonds. Anyone with speculative rating and collateral to secure jumped in. Guess who is coming home to roost ?. To make matters worse, the implosion of CDOs, which bought into majority of leveraged loan issuance in the 2005-2007 period, has significantly depressed market appetite for loans.

If relative credit worsening (CCC and B, relative to BB) is in line with 2002, speculative grade default rate appears all set to reach close to 15% in 2H09. Despite that, high yield paper has rallied strongly this year, off of greatly depressed valuations. With US loans and bonds up close to 30% this year, and equities pricing in a meaningful recovery in earnings, it’s about time to take stock of the asset class, between constituents and vs. equities -

• High yield paper will likely continue to outperform equities through 2009 – Equities are already priced for a significant earnings rebound in 2010E. Short of 30%+ earnings growth in 2010E, there is little else to support relative attractiveness of equities over high yield paper. That said, earnings rebounds are strongest in the early phases of recovery and there are more than a few pillars now to support that earnings could exhibit one of the strongest rebounds in history, if consumer deleveraging and unemployment stabilize. The issue if such stabilization could take place before 2010E is an entirely different question though (I’ll deal with this in a subsequent note). For now, let’s just say that despite the strong high-yield performance YTD, risk-adjusted upside potential for high yield paper (still) remains more attractive versus equities, albeit less so than at the beginning of 2009. The potential for any upside however, appears bleak for CCC and lower rated bonds and loans (see Exhibit 2a), both of which are likely to underperform equities in 2H09E.

• Loans may offer better risk-adjusted yields, but across a range of 10%+ earnings recovery scenarios, bonds offer more upside potential. BB and B bonds are not only offering comparable expected all-in yields (after losses and amortized discount), but under almost any realistic earnings recovery scenario for 2010E, they also offer more upside potential vs. leveraged loans and equities. Overall, bonds appear more attractive, assuming earnings exhibit double digits growth in 2010E.

The 2013-2014 loan re-financing risk – Can the CDO void be filled ?. With many investors fearing a permanent impairment of CLO appetite, concern around loan refinancing has been growing. Following a light maturity schedule through 2012, more than $350 billion in leveraged loans come due in 2013 and 2014. Recent amendments are pushing out maturities, potentially turning the 2013-2014 refi concern into 2012-2014 refi concern. That said, majority of amendments this year appear to have focused on covenant relief vs. maturity extension. Despite the gloom, there are offsetting factors in play - I estimate that of the more than $350 billion due in 2013 and 2014, about a third could potentially face pre-emptive default or get rolled into bonds – Note that about 60% of high yield bond issuance this year went towards loan pay-downs. A range of other options exist – Banks can finally begin holding more of their originations, hedge funds and other investors might return, and profitability might improve from current levels, among other scenarios.

Loan amendments are pre-empting covenant violations. If average maximum debt/EBITDA covenant (of about 5x) of issuance between 2005-2007 were to be enforced, about half of loans originated over this period would be in violation before the end of this year (based on a 30% decline in EBITDA over 2008-2009 period). However, a spike in amendments is allowing issuers to preclude potential violations – The number of leveraged loan amendments year-to-date have already surpassed all amendments between 2006-2008 (see Exhibit 1a). To put things in perspective, the number of amendments this year alone account for about 40% of all issues included in the Barclays High Yield Loans Index. Even as these amendments have increased loan downgrades (see Exhibit 1b - Notice the spike in % of CCC rated loan issues since November last year), they have certainly provided issuers with some respite, assuming economic climate doesn’t worsen in 2010E.

Exhibit 1a - Leveraged Loan Amendments



Exhibit 1b - % of Issues rated CCC


Source: Fitch Ratings, Barclays Capital Fixed Income Indices


Improved high-yield pricing and tax relief have promoted loan-refinancing and exchanges. Relatively improved appetite for high-yield bonds, heightened need on the part of debtors to disentangle from restrictive covenants, and a steeper yield curve are attracting many issuers to replace existing loans with new notes. By some estimates, about 60% of high-yield bonds issued this year have been used to refinance bank debt - It is likely that the ongoing refi activity would have marginally depressed current bond default rates.

Relaxed debt repurchase accounting rules have also contributed to improved sentiment - Under the 2009 Recovery Act, tax on cancellation of debt income, arising from debt repurchases, may be deferred until 2014E, and original issue discount used in debt restructuring is now deemed deductible.

All-in yield for leveraged loans is likely to be better than high yield bonds, but bonds offer more upside across most recovery scenarios. 2009 YTD recovery rates (based on 30-day post-filing quotes) for loans have been in the mid 50s, well below the close to 70% recoveries noted during the 2001-02 period. Further, 1Q09 recovery ratings at Fitch also point to recovery estimates of below 60%. However, even at close to mid-teens default rate and about 50% recovery, I estimate that all-in yield (after amortized discount and losses) for leveraged loans will be about 3 pps ahead of high yield bonds (see Exhibit 2a). However, while loans offer better risk-adjusted upside in a muted earnings recovery scenario, across most other scenarios, high yield bonds appear to offer the most upside potential (see Exhibit 2b).

Exhibit 2a - Est. All-in yields*



Exhibit 2b - Est. Upside potential across earnings recovery scenarios**


Source: Barclays Fixed Income Indices, First Call, Internal Analysis & Estimates
* After expected losses and amortized discount
** Based on 2010E earnings recovery scenarios

High yield notes and loans have traditionally outperformed equities following downturns and are likely to do one better this time, following one of the worst years in 2008. High-yield bonds not only outperformed equities throughout the last downturn (2000 through 2004), but they also handily outperformed in 91, 92 and 93, following underperformance in 90. 2009E could very well end up outperforming 91’s high yield returns of close to 40%.



DISCLAIMER: The information, opinions, estimates and projections contained in this post were prepared by me and constitute my current judgment. The information contained herein is believed to be reliable and has been obtained from sources believed to be reliable, but I make no representation or warranty, either expressed or implied, as to the accuracy, completeness or reliability of such information. I do not undertake, and have no duty, to advise you as to any information that comes to my attention after the date of this post or any changes in my opinion, estimates or projections. No part of this post can be reproduced without permission, unless reproduced with due credit provided for the source. Investment research is provided for information purposes only and does not constitute investment advice or an offer or solicitation to buy or sell any designated investments discussed herein. Please discuss with your investment advisor before investing.

ADDITIONAL DISCLOSURE: As of the date of this post, author had positions in high yield bonds and floating-rate debt.

Cumulative Performance [through mid-July 2009]

Exhibit 1 - Cumulative Performance against benchmarks


Note: CS Tremont All-Hedge performance for 1H of July 09 is not available.


Disclaimer: Performance figures are for informational purposes only and do not constitute investment advice or an offer or solicitation to buy or sell any designated investments discussed herein.

Can India’s emerging low-cost vehicle segment come to the aid of lofty car ownership forecasts ?

Many market participants actively buy the “India Shining” thesis and mindlessly paint it across consumption categories. Multiple domestic passenger car sales forecasts were predicting close to mid-teens growth over the next two decades. Most of these weren’t even factoring a concomitant drop in household savings, a mathematical improbability. Many of these forecasts were published even before the Nano concept was unveiled, which makes their credibility highly questionable. However, the emergence of the low priced car segment couldn’t have come at a better time for forecasters who were banking on an almost seven-fold increase in passenger car PARC between now and 2025E. That growth however comes with considerable risks, given embedded leverage (more on this later in the note) in auto demand - Note that while India’s growth slowed from about 9% to ~7% in 2008-09, new car sales collapsed and reported a flat year in 2008-09, following on +12% growth in 2007-08.

The following are the likely reasons for rosy car ownership forecasts -

1. Expectation that income growth will drive affordability, which it will.

2. India’s per-capita car ownership is very low (see Exhibit 1b) and is bound to increase. Note how high gasoline prices (relative to per-capita income) and low penetration of paved roads explain low ownership (see Exhibit 1a).

3. Expectation that consumer savings can be monetized, since India’s savings rate is too high, relative to other emerging markets, and higher than where major car-ownership “take-offs” occurred. Also, dependency ratio is expected to continue to decline.

Exhibit 1a – Car ownership vs paved roads and gasoline cost



Exhibit 1b - Car ownership vs. per-capita GDP















Source: World Bank, Nationmaster, IMF
Note – Per-capita gasoline cost is based on annual consumption of 300 gallons; Bubble sizes indicate cars/1000 people


In my view, while the above are all relevant reasons, they are 1. Influenced by past performance, which benefited from a significant increase in auto loan penetration, 2. Fail to take into account economic affordability (vs. levered affordability) outside the top quintile of households, and 3. Fail to recognize that Indian household net-worth, disposable incomes and inadequate government social support warrant the need for high household savings.

Even Nano is out of “economic reach” for about 80% of Indian households. Chief constraints to increasing car ownership in India include inadequate road infrastructure, sky-high fuel prices (absolute and relative to per-capita income), and low net-worth level, considering already low disposable incomes (see Exhibit 1a and 1b). Unless households are willing to either considerably cut or completely do away with their current run-rate savings rate (I don’t see that happening – See the exhibit set titled Net-Worth to Personal Disposable Income – a cross-market comparison), a vast majority of Indian households remain priced out of even the low priced car segment. Before Nano came to market, the on-road cost of base versions of some of the cheapest models on offer (Maruti Suzuki Alto, Chevrolet Spark, Hyundai Santro etc.) was northwards of $5,000, effectively pricing out a material number of Indian households, even within the top decile, let alone the top quintile. By my estimate, the 90th percentile Indian household falls just short of world average Net-Worth/PDI ratio – In an economy with a very low level of absolute disposable income and inadequate social security net, the bottom 90% is simply not in a position to significantly compromise their monthly savings by indulging in big-ticket discretionary spends such as car purchases. Consider this – India spends significantly less on social expenditure than even many poor neighboring South Asian economies. Both Bangladesh and Nepal, for instance, spend considerably more on social expenditure than India does. Further, any material drop in savings within the top quintile of households will significantly reduce India’s overall household savings rate, given that the top quintile accounts for almost half of Indian household savings.

In my opinion, absent the low-cost car segment, car ownership will largely remain confined to a subset of the top household quintile in the near-term (The 80th percentile household will likely not be able to afford an ex-Nano base version until 2016E, assuming they are comfortable with seeing their savings rate getting slashed by at least a third). With it, car ownership can spread across the top two quintiles by 2025E (see Exhibit 2). Nano may not be your car but you cannot seriously buy into the estimates that project seven-fold increase in passenger car PARC in India between now and 2025E (@12% average unit growth) without considerable penetration of the low priced passenger car segment.

Exhibit 2 – Plotting passenger car PARC against affordability


Source: Indicus, National Registrations Data, Internal Estimates & Analysis


Shoving capital down consumer throats. Not only is a vast majority of India’s car sales financed, but it is also done at particularly high LTV, considerably above global averages (Before credit crisis forced more stringent underwriting, average LTV of 90% or more on auto loans was pretty common, which has since come down to closer to 80%; EMI to Income of 50% is still viewed as “comfortable”). With onerous seizure and recovery policies, auto loan growth now desperately needs a breather.

While capital availability grew comfortably ahead of affordability over the last decade, that trend will almost certainly not last much longer (given already high % of financed cars), suggesting that the dominant driver of car ownership, going forward, will have to be real per-capita income growth, which is unlikely to sustain average growth of more than 5%-7%.

Be cautious on contribution from replacement cycle. While credible statistics on Indian vehicle scrappage don’t exist, it’s fair to assume that Indian passenger car scrappage rate is likely to be a very low single digit figure (There is a laundry list of reasons) - The average age of existing Indian car fleet is likely in the neighborhood of about 5 years, or about 50% of US, and average odometer reading for a 5-year old car in India will be no more than 40% of the reading on a same age car in the US. Given significantly lower use and affordable service and aftermarket component costs, scrappage rate is likely to be significantly less than the mid-single digit rate in the US.

Most Asian car ownership “take-offs” began at significantly higher per capita income levels than where India stands. Car ownership “take-offs” in Japan (1960s), South Korea (1980s), Taiwan (1970s) and China (present) occurred when per-capita incomes (in 2000 constant prices) in these markets were 125%-290% ahead of where India is today (see Exhibit 3). Also, in each case, gas prices were far more manageable relative to household income than they are in India now. That said, Indian car ownership take-off is benefiting from globalization like no one (barring China), did. More importantly, India’s savings rate is well ahead of where take-offs occurred in the above cases, providing some room for it to go down, which has to occur to support 12%+ CAGR from here on, unless India reports unprecedented real income and employment growth, or over-generous financing miraculously gets sustained for a decade. In sharp contrast, Japan, South Korea, and Taiwan, reported a sharp increase in savings rates during take-off periods, even as car ownership spread rapidly.

Exhibit 3 – Asian car-ownership “take-off” decades – Beginning GDP per-capita vs. per-capita sales growth


Source: GM, CSO, Internal Estimates & Analysis


(In)sanity Check - Segregating the top quartile as a separate economic group. If top 25% of Indian households were to be addressed as a separate fictional economy, Indian gasoline pump prices remain unchanged through 2025E, and passenger car sales were to average 11% growth over this period, I estimate that such an economy would have car ownership of about 250 cars for every 1000 people in 2025E. This is based on a fairly aggressive assumption that real per-capita income growth for the top 25% of households will average close to 8% through 2025E. The resulting quadrants (car ownership against retail gasoline prices and per-capita GDP (2008 constant prices)) are shown in Exhibit 4. The results indicate that even 11% average growth expectations appear aggressive, at least when compared with the peers on that plot – While our fictional entity will have a per-capita GDP (in 2008 constant prices) about 15% below peer average, it’s resulting car ownership would be 15% ahead of peers, second only to Malaysia on that exhibit, a country whose per-capita GDP is 16% ahead of our fictional entity and retail gasoline prices more than 40% below Indian pump prices. Comparison using diesel-prices wouldn’t yield anything that is materially different. Any way one looks at it, the mid-teens CAGR that some forecasts suggest is likely based on past performance, assumes aggressive acceleration in wage growth, and largely ignores an almost certain and profound impact on household savings, given high car ownership costs.

Exhibit 4 – Addressable top 25% of households – Est. car ownership vs. GDP per capita and gasoline prices



Source: Misc, Internal Analysis & Estimates
Note – Bubble sizes indicate cars/1000 people; Disposable Income multiplier was used to calculate GDP for top 25% households.


Can India repeat its 12% CAGR performance of the last decade, which was fuelled by generous loan underwriting ? – I wouldn’t hold my breath, but a sustained strong growth of new low-priced passenger vehicle segment will almost certainly be required to emulate past performance. In addition, an unlikely scenario of significant monetization of household savings will need to play out. The median Indian household in the top quintile makes just enough to save about a fourth of his disposable income after purchasing a model such as Alto, Spark or Santro. Those in the bottom fourth of the top quintile will see their savings rate dip into mid-high teens if they were to purchase any of the above mentioned models (Such households typically have less than 4 years of their disposable income as savings, which is a low denominator to begin with, and close to half of savings are tied up in physical assets). On the other hand, purchasing Nano’s air-conditioned variants is much more affordable for households at the bottom of the top quintile, with savings rate dipping into mid-20s post-purchase. In either case, it is critical to appreciate the fact that impact of car ownership costs on household savings across a vast majority of Indian households in the top 2 quintiles is too striking to ignore. Given where Indian household net-worth stands, such a change can only take place gradually and places economic limitations to car ownership . More importantly, if 12%+ market growth is to materialize, I estimate that the low-priced car segment (which only includes Nano at this point), will account for at least a fifth of the increase in Indian car PARC between now and 2025E.



DISCLAIMER: The information, opinions, estimates and projections contained in this post were prepared by me and constitute my current judgment. The information contained herein is believed to be reliable and has been obtained from sources believed to be reliable, but I make no representation or warranty, either expressed or implied, as to the accuracy, completeness or reliability of such information. I do not undertake, and have no duty, to advise you as to any information that comes to my attention after the date of this post or any changes in my opinion, estimates or projections. No part of this post can be reproduced without permission, unless reproduced with due credit provided for the source. Investment research is provided for information purposes only and does not constitute investment advice or an offer or solicitation to buy or sell any designated investments discussed herein. Please discuss with your investment advisor before investing.

ADDITIONAL DISCLOSURE: As of the date of this post, author owned depository receipts of Tata Motors (NYSE:TTM)