Sorry!! The article you are trying to read is not available now.
Thank you very much;
you're only a step away from
downloading your reports.

Analyzing the QE Meltdown, Ex-Ante


Correlation does not imply causation.

MINYANVILLE ORIGINAL Last week I attended an investment conference in Memphis hosted by FTN Financial that catered to its vast depository institution clientele. The overriding theme of the presentations was how to manage your bank's portfolio in this low interest rate environment. I didn't agree with everything I heard, but overall, it was a very value-added experience.

After a night of sucking down beer and Rendezvous ribs, FTN's interest rate strategist Jim Vogel delivered a presentation titled Understanding Low Interest Rates, Supply and Demand for Credit in the US.

For the most part, it was the same story we've all heard about the debt bubble. Vogel goes through the basics of how we got to this point of low interest rates and then addresses whether this is a cyclical change or a structural one.
  • Real interest rates should be negative to reflect aversion to economic leverage and poor risk-adjusted results on other asset classes. Real rates follow debt demand, and the Fed isn't spurring appetites for new loans.
While I didn't disagree with his point of view and analysis of the deleveraging cycle, I jotted in my notes confirmation bias. He basically gave the room of bankers a good excuse for owning and buying the negative coupons his firm has been selling them because the market was rational.

This was an exercise in ex post analysis. Vogel's conclusion was based on what had already happened. Like in the bullet above, explaining the demand for negative coupons was based on the results of returns on other assets. Right, but that's after the fact.

As an investor, I don't want to know why something is priced the way it is today. I want to know what is going to change price in the future. You can't define the risk/reward of buying and holding an asset based on what has caused the price you are paying; you have to define it by what variables will drive the price going forward. In order to anticipate price, you have to identify what risks are not currently discounted but what could be discounted.

Wall Street and Fed policymakers can't handle this concept, and that is why they are prone to blowing up your portfolio. Their models are a function of ex post analysis. They extrapolate data that has already been reported into the future based on past trends or similar past relationships. However, one of the biggest mistakes investors make is interpreting cause and effect in the market. It's easy to draw correlations to explain market behavior, but just because something appears to be related or was in the past does not mean that it is or will be in the future. Correlation does not imply causation.

During the Jackson Hole symposium over Labor Day weekend, Columbia Professor Michael Woodford generated a lot of buzz by proposing that the Fed should look at a policy of targeting nominal GDP. This was a concept we had addressed a month prior. On July 31 in Bernanke's Astonishingly Good Idea, I posited that despite Bernanke previously dismissing it as an option, he was getting desperate and that a nominal GDP target was on the table.

I think we will start to hear more about a program along the lines of a nominal GDP (NGDP) target that is used to justify direct purchases of private assets.

The idea of targeting NGDP would be to place an extrapolated dollar amount on the output gap based on a specified growth rate and the Fed would commit to tightening or easing based on where nominal growth deviated from that trajectory going forward.

I didn't just lay out the possibility of a NGDP target initiative, but went a step further, framing the consequences and potential market reaction.
If the Fed did go nuts and drop money from the skies, they would be naïve if they thought it would not elicit a severe market response. The inflationary tail risk would likely manifest itself in an immediate collapse in the US dollar and a corresponding rise in long term interest rates.

This is the essence of ex ante analysis. I took a concept that was not discounted by the market but could become a real possibility given the state of the employment situation. Instead of finding a reason to be bullish because the Fed would be buying, which was the consensus view among market participants, I focused on why this would be bearish for a market that was already very long.

Last week in The Most Important Market No One Is Watching, I analyzed the price action following Friday's weak employment numbers as indicative of a market that was fighting for positioning and concluded that the result suggested the market was not concerned about the weak growth, but rather was preparing for the inflationary implications of QE III.
Since the miserable June NFP that only saw 45K jobs added, the contract (Dec now front month) has been confined to a range with the big pivot at 150-00 which is the same level that was tested in Thursday and Friday's intense volatility. This is not a random coincidence. The market is battling in here.

Friday the action in the yield curve told the whole story. The belly, 5-year and 7-year led the performance finishing better by 3bps and the 5-year/30-year spread finished 6bps steeper on the day. The steepening bias in the curve was a clear message that the bond market is preparing for a higher inflationary discount as the result of QE III.

Again, this is ex ante analysis applied. The casual observer looks at the weak data and 1.67% 10YR yield ex post and assumes the market remains fixated on what was a continuation of recessionary economic data. I took the approach that by virtue of the steepening curve, the market viewed the data as so weak that it would force Bernanke's hand into aggressive action, which translates into a higher inflation discount and thus higher long term interest rates.

Regardless of whether QE III is an implicit target of the output gap or an explicit one, the goal is the same. Bernanke knows employment won't come down with the output gap in place and it doesn't seem he will stop until he closes it. There is no doubt that the inflationary path to close the gap produces a steeper yield curve and higher long-term interest rates.

Coming into last week, the bond market was on alert and supply was coming, specifically in the long end with a 10YR and a 30YR auction right in front of Thursday's FOMC announcement. With the bond contract settling out the week below the critical 150-00 level, the risk was that the market wanted a test of the lower pivots I had also identified.
If the trading range that has confined the contract since June breaks to the downside, the pressure could be on. In addition to the crucial 150-00 level, I have lower pivots of 148-00 and 146-16. These levels should be respected if tested and I would not fade the move from these big pivots in either direction.

The 10YR auction on Wednesday was sloppy with weak participation from both direct and indirect bidders. The market was testing 148-00 already down 1.5 points on the week. But despite the poor auction, the contract found support and consolidated this lower pivot. After a considerable concession Thursday the market absorbed the 30YR auction fairly well and remained near the 148-00 level as we approached the FOMC witching hour.

The FOMC release hit the tape and announced that QE III was forthcoming in the form of increased MBS purchases, but that wasn't nearly as important as the following statement:
If the outlook for the labor market does not improve substantially, the Committee will continue its purchases of agency mortgage-backed securities, undertake additional asset purchases, and employ its other policy tools as appropriate until such improvement is achieved in a context of price stability.
This was a very dovish development and meant QE was now open-ended, and not predicated on a time frame but on an economic outcome. Just as we suspected last week, it was essentially an implicit targeting of the output gap. The bond contract immediately collapsed and sliced right through the 146-16 pivot, but quickly bounced back and stabilized with only minimal damage. Still, while mortgages screamed higher, the reaction out of Treasuries was clearly focused on the inflationary implications.

The market had absorbed a lot of supply and a surprising launch of QE III so you would have expected Friday to be a slow day. The overnight session, however, proved to be a different story. The curve steepening had continued and the bond contract was gapping lower through our 146-16 lower pivot and had traded as low as 145-16 just before pit trading opened.

One thing that was not on the agenda at Thursday's FOMC meeting was NFL football, but it probably should have been. On Thursday night, the Bears played the Packers and the traders in the pit were no doubt out late, hung-over and in a bad mood when they came in Friday to find stops triggered and bonds gapping lower. I can assure you this has an impact on market liquidity.

I don't know this for sure, but my guess is that the locals were not there to step up and take down a wave of sell orders. They probably were out to get in front of the selling not to take the other side. The bond contract opened down almost 3 points from Thursday's close to 144-16, and after a brief oversold bounce to 145-16, rolled back over to finally settle just under 145-00. Nevertheless, severe technical damage had been done with both prices and yield closing out the week at levels not seen since mid-May.

Bloomberg reported after the close:

Treasury 30-year bond yields (^TYX) climbed the most in three years after Federal Reserve Chairman Ben Bernanke's pledge to keep adding stimulus even as the economy improves spurred concern inflation will accelerate

The yield gap between 10- and 30-year Treasuries widened to the most in a year yesterday amid bets the value of longer-term fixed-income holdings will erode more over time.

Yields on 30-year bonds increased 26 basis points, or 0.26 percentage point, this week in New York, according to Bloomberg Bond Trader prices. It was the biggest jump since August 2009. Yields touched 3.11 percent, the highest level since May 4. The price of the 2.75 percent security due in August 2042 slumped 5 2/32, or $50.63 per $1,000 face amount, to 93 13/32.

The 5YR/30YR spread we cited last week had gapped out and had virtually wiped out all the flattening that had occurred from Operation Twist. The bond contract closed 1.5 points below the second lower pivot we cited on Monday and the 30YR yield appears to have broken the down trendline that has been resistance since last year when yields were over 4.00%.

When QE II was announced in 2010, the curve violently steepened. The 5YR/30YR spread widened 90bps between the widest points of both August and the November with the 30YR yield rising from 3.50% to 4.75% in February 2011. Here's where it gets even trickier: That was with much higher coupons. The August 2010 30YR coupon was 3.75%, but this 30YR coupon is 2.75%.

As you know, for a given duration, the lower the coupon the more sensitive price is to a move in interest rates. The 3.75% 30YR coupon has a duration of 16.7 while the 2.75% 30YR coupon has a duration of 19.1, meaning that the 2.75% is much more sensitive to a move in interest rates. This was a fact not lost on the market last week where the extended duration can be attributable to over 10% of the total move.

This may not sound like a material amount but when you amplify this throughout the entire bond market and include the negative convexity in MBS and Agency debt you have the recipe for quite a toxic environment if this market is ready to turn.

Late Friday, Egan Jones exhibited ex ante analysis and downgraded US Treasury debt by one notch to AA- due to the inflation risk as a result of QE III. Many in the street brushed this off as irrelevant, but I think Egan Jones raised a valid point and his reasoning is in line with a risk we addressed in The Bond Market's China Syndrome in early August.
Today... the conditions are in place for a similar rise in yields. In addition, with long duration coupons substantially below their level in 2010, a move just back to where yields were before last year's August equity market crash would produce a massacre in the bond market, potentially leading to a dangerous self-fulfilling meltdown.

The US fiscal situation is no different than Europe, and you can bet that if our yields start rising, our credit condition will rapidly deteriorate. This will feed on itself. Suddenly sentiment will shift and our reliance on foreign financing will be at risk as the reserve currency status is called into question.

So we have a situation here where there is a tradeoff between inflation, employment, and interest rates. It's not clear to the market what level of inflation is acceptable to Bernanke to bring down unemployment, but it is clear that the market is prepared to respond and there is an interest rate limit that would threaten US solvency. What is max pain for the US Treasury? I don't want to find out, but maybe Bernanke does.

The ex post analysis of QE III has been focused on the Federal Reserve keeping interest rates low for a very long time, concluding that bond yields will also remain low. The ex ante analysis has been focused on the implications of the inflationary discount in the long end of the curve and concluding that the conditions are in place for a nasty spike in interest rates that could manifest itself into a full blown meltdown.

I'm not calling for a bond market crash and maybe Bernanke can hold it together, but last week was a shot across the bow. I've said it many times before: Bernanke is prone to blowing up his own trade -- and if he's not careful, he may blow up the biggest one of all.

Twitter: @exantefactor
No positions in stocks mentioned.
Featured Videos