What is the historical link between rising bond yield and credit spreads?
Historical empirical evidence shows that credit spreads tend to peak at the end of recessionary periods. Then, credit recovers along with rising government bond yields, as bond markets usually start to price in higher real growth rates, increasing inflationary risk and a change in monetary policy. The table below, which covers the cycles of the last forty years in the United States, confirms that a credit rallies following recessionary phases have always been associated with a substantial rise in bond yields. Furthermore, post-recession credit rallies are more vigorous and run out of steam sooner when bond yields rise sharply, as occurred in the 1980s and 1994.
A negative correlation between spreads and yields in post-recessionary phases does make sense: recovery means improving credit metrics through increasing cash flow generation, though the best of debt reduction is usually already past at that time. Rising inflation expectations, which push yields higher, also mean better pricing power prospects for companies, and initial rate hikes by central banks usually tend not to harm equities and corporate bonds. Furthermore, spreads offer a cushion to the negative duration effects arising from an increase in bond yields. Therefore, investors still perceive corporate bonds as attractive compared to government bonds in these phases.
Is this also the case in the present cycle?
The first graph shows the link between Treasury yields and BBB-rated corporate bond spreads in the latest cycle from before the GFC up to the present day. As in previous cycles, the usual link occurred: spreads peaked during the final part of the recession and then rapidly tightened in the recovery that followed, spurred and sustained by the unusual monetary and fiscal policy measures. The graph shows that the post-recession recovery phase is long past and that a prolonged phase of normalisation has taken place since 2010. The European crises (the Greek crisis in 2010 and the more widespread sovereign crisis in 2011) contributed risk-off phases in which spreads and bonds moved consistently with the usual link. A period of normalisation took place until the “tapering phase”, which had the usual link, but spreads had very limited tightening capacity when yield rose, as in the “tapering” phase shown by the bubble in the graph. In simple terms, the best of the credit cycle is behind us (in terms of spread tightening and excess return delivered on government bonds) and a normalisation of credit risk premiums has already taken place, especially relative to bond yields: the usual link may no longer be the most likely. At the same time, growth, inflation, inflation expectations and the prospective tightening cycle all look less extreme compared to previous cycles, and this should lead to a milder upward trend in bond yields.
Where are we in the tightening cycle?
In recent weeks, the FOMC confirmed its plan to hike Fed funds rates within 2015 but expressed it would proceed very slowly thereafter. The FOMC’s statement changed only slightly at the June committee; however, it indicated a certain degree of confidence regarding the economic outlook, particularly the inflation outlook, following the first signs of wage acceleration (which still needs to be confirmed). However, the FOMC officials lowered their Fed funds projections (the “dots”) for the fourth time in a row: the median projection stands at 0.50/0.75% for end-2015, 1.50/1.75% for end-2016 and 2.75/3% for end-2017.
The Fed fund projections (“dots”) should be interpreted with caution. On 5 June, William Dudley, the president of the New York Fed, explained that FOMC members would put a very wide confidence interval around their projections if possible. It is striking that the dots gradually converge towards the Taylor rule referred to in numerous speeches by FOMC members in recent months (see, for instance, “Normalizing Monetary Policy : Prospects and Perspectives”, Janet Yellen, 27 March). The dots are now in line with this Taylor rule for end-2015 and end-2016 but are still above for end-2017 (the Taylor rule would indicate Fed funds at 2%). The dots will probably continue to decline over this timescale. We anticipate only one increase in Fed funds in 2015. As Janet Yellen indicated during the press conference, the Fed funds tightening will be very gradual overall: the terminal rate of the Fed funds will be low at the end of the cycle and it will not be reached before 2018.
What are the consequences for the US bond yields? Generally, periods of monetary policy tightening coincide with a flattening of the yield curve. Like Fed funds tightening, the flattening of the curve should be very gradual in the coming quarters. Long-term yields are already relatively high in the US. At the end of tightening cycles, long-term yields (5 and 10-year maturities) are generally close to those of the Fed funds: if the terminal Fed funds rate is assumed to be 2.50%, there is not much room for a rise of long-term yields. The short end of the curve is more likely to rise. This being said, one cannot exclude that the long-term yields will overreact to the lift-off, or that the Fed will take more time for the Fed funds tightening process.
The recovery cycle we have been watching since late 2009 is, from a purely fundamental point of view, quite unusual
Growth in US earnings since the end of the recession in 2009 has been driven much more by improved margins than by revenue growth. Last year, US corporate margins hit an all-time high as a result of significant cost compression – specifically on labour costs. In fact, there has been virtually no increase in wages since the crisis ended. However, the employment situation has improved significantly. More than 200,000 jobs were created over the past 12 months, sending unemployment down to 5.5%. The reason for this dichotomy is that these jobs have been created primarily in low-wage sectors (food service and health care), and activity has been at an all-time low (63% at end-May 2015).
In recent years, this flat demand has resulted in weak investment spending. Nonetheless, the leverage of US businesses has increased in the past few quarters, now standing at the midpoint of the re-leveraging cycle. US businesses have benefited greatly from nearly a decade of ultra-accommodative monetary policy from the Fed. The funds raised on the financial markets over the past three years have hit record levels. We can only wonder – legitimately – about the use of the funds raised versus the weakness of investment spending. In a weak-growth environment, businesses have opted for mergers and acquisitions and share buyback transactions.
Many indicators are converging to announce an upcoming wage acceleration, which would be quite consistent with an increase in the fed funds rate. This wage increase would strengthen demand, which would lead to a more pronounced investment recovery. Business leaders need greater visibility on demand before committing to irreversible investments. It should be noted that the first fed fundshike is happening in a context where corporate debt is already high. Mergers and acquisitions and share buybacks have returned to cycle highs.
It is not a given that the first fed funds hike will come with improved business fundamentals, as in previous cycles. Nor do we expect US business leverage to erode substantially, given the weak recovery.
Bond yields and investment flows into HY
The link in recent years between flows and yields looks stronger in the US than in Europe: sudden and sharp upward trends in rates, affecting total HY returns, tended to trigger subsequent strong outflows on the other side of the Atlantic. Since 2013, and especially since the “tapering” announcement, the link between investment flows and bond yields has also become stronger. The graph confirms that in each of the last three years, upward moves in bond yields tended to trigger considerable outflows from institutional investors to US HY bonds. Not only the level but also the volatility of bond yields has increasingly become a driver of outflows from investors in high-beta credit products. Sudden and significant negative effects produced on total returns by higher bond yields after many years of remarkable positive performance are among the factors behind this link. Looking forward, this link is here to stay and will be a function of future moves in yield curve slopes and levels. In this respect, and according to our view on anticipated moves in yield curves, a bear flattening scenario is likely to impact more heavily on short-to-medium maturities on the HY bond market rather than on longer maturities. The comparison with dedicated investors in European HY bonds is quite telling as well. Inflows into European HY bonds and ETFs were and still are much less influenced by yields and volatility, mainly due to two factors. ECB QE will last at least until September 2016, and the phase of the cycle is precisely the one in which higher bond yields go hand in hand with tighter spreads: in a nutshell, the initial recovery phase. Even in the most recent period of global bond re-pricing, and contrary to US trends, the sudden and dramatic upward move of Bund yields failed to spur outflows from European HY funds and ETFs.
A bear flattening move has the potential to impact more heavily on short-tomedium maturities, as occurred in summer last year. Therefore, we would be more cautious on this segment of US credit curve. Credit quality is another important factor to be taken into account when the link with flows is considered, in light of recent years’ experience. In summary, we favour high-quality names with mid–to-long-term maturities in the US.
Spreads tend to peak at
Spreads have normalised
The link in recent years