A wise old trader once told me that the change in trend often happens at the roll. For the non-futures traders, the roll is when the back calendar month “rolls” to the front traded contract. For example, going into last week, June was the front month in Treasury futures, and on Friday we rolled into September.
A change in trend can occur when you have a large move into the roll because many traders may elect not to roll positions. In the case of Treasuries, the trend into the roll has been a violent move lower, however the sellers of June may not re-short September and thus cancel out the position on the roll, in this case covering, rather than rolling into the next month. This countertrend demand (supply) in effect absorbs the supply (demand) and alleviates the pressure. With multiple days of open interest turnover in the bond futures, coming into the roll, there has no doubt been a change in ownership. With Friday’s massive reversal on elevated volume, this may have been the final flush reversal needed to change the trend.
Regardless, the bond market volatility has been intense, and there is a lot of blood in the street. Most would presume this volatility is a direct product of the Fed looking to taper QE, but in reality, it is the buying program itself. The bond market has accommodated the Fed up until this point, but the negative interest rate regime has not been a costless exercise.
On May 13 I wrote the following in The Dark Side of QE: The Yield Curve Adjustment Process
What I think gets lost in the discussion about the cost/benefit of QE, and more importantly, how long it has gone on, is the fact that there are billions of potentially very volatile long duration low coupons on the balance sheets of investment portfolios. Bond math dictates that for a given duration, the lower the coupon, the more volatile the price. A 2% 10-year is much more price-sensitive than a 4% 10-year. Due to the nature of these low, long duration coupons, the adjustment process is likely to be very chaotic, making last week look like a game of tiddlywinks. This is bound to have a profound impact on portfolios and on sentiment.
This is exactly what’s going on. The selling and volatility has been intensified by the negative convexity of low coupon, long duration of mortgages. This negative convexity forces hedged MBS investors to add duration as bond prices rally and sell duration when prices fall, often exacerbating the move. The current coupon 2.50% Fannie Mae
(OTCBB:FNMA) 30-year mortgage is trading to yield 2.90% and his up 50bps since the Fed committed to buying virtually all of new production.
After the QE announcement the MBS spread to the 10-year collapsed to near-zero, yet the subsequent sell-off has seen the spread steepen back out 75bps from low to high and is now back toward more normal historic levels. This convexity blowout is no doubt responsible for the volatility in bond prices, but it is the not catalyst for the move. This is not the first time higher bond yields were being misinterpreted for an accelerating growth discount, and due to the nature of the move today I think we have a similar misinterpretation.
We have been told by a lot of people who don’t carry a bond position that the market is overreacting to the tapering rhetoric and that bond market participants don’t understand monetary policy. Ironically, contrary to what the financial media and pundits are asserting, this move in interest rates has not been a reaction to improving economic data but rather to falling inflation. This move higher in yields has been led by real interest rates. While bond yields have risen dramatically over the past month, yields on TIPS (Treasury Inflation Protected Securities) have exploded to the upside.
Five-Year and 10-Year Nominal Vs. TIPS Yield
Last week in Bernanke’s Worst Nightmare: Rising Real Interest Rates
, I broke down the bottom line:
The bond market was under significant pressure with a renewed focus on the eventual Fed taper/exit from QE III despite repeated attempts by central bankers to convince participants otherwise. Many strategists and pundits in the media tried to convince investors that just because the Fed would be tapering asset purchases didn’t mean that they would be tightening and that stimulus would remain in place for the foreseeable future. This misses the point.
The bond market doesn’t care about how much the Fed buys per se. Whether QE is $4 billion per day or $2 billion per day is irrelevant in a market that trades $800 billion (Treasuries and MBS). With the Fed simply backing off from a full commitment to manipulate the yield curve, the bond market will turn its focus on pricing a natural equilibrium rate of interest. This is the essence of the yield curve adjustment process, and recent price action is just a taste test of how this will unfold.
The focus of most participants is on the rise in nominal yields, but the story is being told by the move in real yields.
The dark side of QE is the idea that no one really knows where the natural rate of interest really is, and finding that equilibrium is going to be a very uncomfortable adjustment process. Regardless to what extent you believe the Fed has influenced interest rates, the fact is we still don’t know where the curve should
trade absent QE. However, with the move in real interest rates leading the adjustment, we can make some assumptions about where real and nominal could settle.
Friday we got the first Q2 reading of April personal consumption expenditures, and the PCE deflator and the decelerating trend since Q4 2012 remains in place. Though admittedly highly lagging, these are two critical data points because they represent the largest component of nominal GDP (consumption) and the Fed’s preferred bogey for inflation (PCE deflator). April YoY personal consumption grew at 2.8% following 3.3% growth in March and 3.6% in December. The PCE deflator registered 0.7% after 1.0% in March and 1.5% in December.
PCE Vs. PCE Deflator
While Friday’s employment report may show continued improvement in the labor market, the trends for growth and inflation are clearly lower and are within a recessionary range (shaded). Both the 2.8% growth in consumption and the 0.7% inflation are among the lowest growth rates in 50 years. It is this rapid deceleration of growth and inflation that are actually driving interest rates higher, and this is a potentially troubling development for asset prices.
TIPS are indexed to the CPI and the spread between nominal yields and TIPS yields is the market’s discount for future CPI inflation, also known as the “breakeven” inflation rate. The spread between the market’s discount for inflation and the current reported inflation is north of 100bps and ex the 2009 blowout, is the widest in the last 10 years. Essentially this means that the market is currently assigning a very steep inflation premium relative to what inflation is actually paying TIPS holders. I think this is in the process of changing. Assuming the inflation trend remains in place, this spread is likely to narrow significantly -- especially as the Fed backs out of the QE III program. As this CPI breakeven spread converges with actual CPI inflation, it will foster further compression in the spread between nominal and real interest rates.
Ten-Year B/E Vs. CPI
This inflation spread narrowing is a bit unusual because it is occurring in the context of rising interest rates. With the exception of 2008, previous cycles of inflation premium tightening were due to falling interest rates, with the nominal yield pushing down on the spread. This inflation premium tightening due to rising real rates is a very different development than both previous rising rate cycles and inflation premium narrowing.
One of the prime targets of QE was the credit market, and one of the biggest beneficiaries of falling real yields has been high yield bonds. The negative real rate regime was a windfall for credit risk because there are no expensive risk premiums when risk-free yields are negative. As I noted in The Low Spark of High Yield Boys
, this performance in HY was in turn directly responsible for equity market multiple expansion which has been the primary catalyst behind the stock market rally:
The casual market observer would look at the parabolic move in the S&P 500 (INDEXSP:.INX) and assume stocks have led the rally in risk, and from a price-percentage-gain perspective, they have outperformed HY bonds. However from a risk premium/multiple perspective HY has massively outperformed stocks, suggesting the rally in stocks is actually a function of multiple expansion on the back of HY credit risk premium tightening. This revelation has important consequences going forward because of where credit sits at this stage in the cycle both on a relative basis and in outright yield.
High-Yield 5-Year CDX Index Vs. 5-Year TIPS
It stands to reason that if the cycle of real interest rates is turning, then every asset that was inflated by these negative yields is at risk. We’ve already seen what has happened to gold and other metals, and I think we are naïve to think that credit and equity would be exempt from this adjustment. We have already seen a crack in HY, and on Friday the spread of the HY CDX index widened 18bps.
The correlation between TIPS yields, HY, and stocks is unmistakable, both fundamentally and technically. Even if you believe that the Fed will continue to stimulate and that reduced QE is not tightening, there is a clear diminishing return which should continue to exert upward pressure on real yields. I think investors should prepare for a real world real rate of 1.0% both in the 5-year and 10-year part of the curve. A 1.0% real yield should coincide with much wider credit risk premiums and much lower multiples on stocks.
S&P Multiple Vs. 5-Year Breakeven
On balance, rising real yields and lower inflation premiums are bullish for the curve, which should trade with a flattening bias. You have seen this action in the 10-year/30-year spread that has flattened with bonds moving in either direction. If real yields move back to 1.0% and inflation continues to moderate near the 1.0% level, then nominal 5-year and 10-year yields should settle in the 2.0% area.
In this normalized yield curve adjustment amidst a low inflation environment the 5-year probably gravitates towards 1.50%-2.0%, the 10-year 2.0%-2.5%, and the 30-year 2.50%-3.0%. Based on Friday’s close, the 5-year is rich, 10-year is fair value, and the long bond is cheap. The consensus looking at nominal wants to be short the long end of the curve, however when analyzed in the context of rising real yields, the curve should flatten, thus punishing the masses hiding out in the front.
We have a lot of different markets that, when analyzed on their own, are saying something different, but when analyzed together, are saying the same thing. The market is unwinding QE, but what is still unclear is whether this is simply the unwinding of QE III or all of QE. This is a critical distinction, and I think over the coming summer months the market is going to make it clear how this cycle is going to play out.
On Friday a technical formation called the Hindenburg Omen was cited on CNBC as a reason for the weakness in stocks. I don’t pay a lot of attention to these superstitious indicators, but I do like the Zeppelin reference in terms of market sentiment. It’s like everyone knows the market is floating on hot air and is doomed for disaster, but they will continue to play the game until the music stops. One of the greatest rock albums in history is Led Zeppelin I, and on the cover is a picture of the Hindenburg going down. From front to back this iconic set of blistering rock and roll is an apt description for today’s paranoid QE driven market. Cue the play list:
"Good Times, Bad Times"
"Babe, I’m Gonna Leave You"
"You Shook Me"
"Dazed and Confused
"Your Time is Gonna Come"
"Black Mountain Side"
"I Can’t Quit You Baby"
"How Many More Times"