HOW TO PUT ON THE STEEPENER
2017-12-27 2pm EDT | #bonds #yield curve
Over the Christmas break, there has been a lot of chatter about this great chart from 13d Research that has been labeled, “the most important chart in the world.”
During the past thirty years, the US 10-year yield has been immersed in a steady decline, with brief run-ups that have coincided (caused?) financial market crises. The 1987 stock market crash, the 1990 S&L crisis, the 1994 bond rout, the 2000 DotCom bubble and the 2007 housing debacle - they were all preceded by a rise in the 10-year US treasury yield.
The trouble with this analysis is that the magnitude of the yield run-up was different each time. The 1990 S&L crisis barely saw a rise, while the 1994 bond rout and 2000 DotCom bubble were much more dramatic. So although this chart is interesting, it doesn’t necessarily help with timing.
Is it the absolute rise that matters? Or is the catalyst the touching of the fabled trend line?
When I tried to recreate this chart, at first I had trouble getting the peaks to tick the trendline.
Then I realized the chart had a monthly timeframe, so I tried again.
That’s the ticket! But the fact that the difference between monthly and daily prices can throw off this analysis should be a warning call about the imprecise nature of this analysis. I am not trying to say it is useless, just that this price action should be viewed only in the longest of timeframes. We are talking months, not weeks or days. So to conclude that because the 10-year is currently ticking against this trendline we should expect a crash any moment is foolhardy.
Yet I can’t help but think 13d is correct in calling this the most important chart in the world right now. I am not sure if this trend of lower highs in yield will continue, but investors should give it their utmost attention as it will most likely determine the fate of financial markets over the next year.
Can the economy handle fewer interest rate increases before rolling over? Does the ever-growing-indebted financial system continue to stall at increasingly lower yield levels? And if so, does that mean we are on the cusp of the next great crisis? Will the Fed have inadvertently tightened into the next slowdown?
Or has something changed? Does the obscene pile of Central Bank balance sheet expansion since the 2008 Great Financial Crisis mean that all bets are off? Is monetarism truly dead? Or were the counter-productive fiscal tightenings of the last decade offsetting all the monetary ease, and now that the foot is off the fiscal brake, are we about to experience the full effects of the monetary madness?
I don’t have the answers, but these are the questions investors should be asking themselves.
And embedded into this problem, lies the fate of the yield curve.
If you are a believer that the economy is over-indebted, ready to collapse at the slightest increase in rates, then you probably are quite sympathetic to the flattener trade.
And to a large extent, this is how market participants are set up. Have a look at the CFTC speculative position in 5-year T-note futures.
The market is net short the most 5-year contracts ever. Yup. This is a historic net short position in the 5-year area of the curve.
But this negativity towards fixed income doesn’t persist as you head further out the curve.
Up until last week’s report, specs were actually long the 10-year sector.
The trend continues with the long bond future pushing up against all-time net long speculative position.
But then, just to confuse us, specs are short massive amounts of ultra-long bond futures.
To try to understand what’s happening, I have created a duration adjusted index of the long bond and ultra-long bond future.
The specs are net short the long end, but it’s nowhere near the all-time short position of the 5-year.
On the whole, specs are most short the 5-year. They are long the belly with a flat 10’s position and a net positive long bond future position, with a strange large short position at the far end of the curve.
But make no mistake. The specs are leaning with flattening positions. If they believed the curve was about to steepen, they would have record short positions in 10s and the long bond futures instead of 5s.
Yield curve signals
What does this mean?
Well, Minneapolis Fed President, Neel Kashkari recently gave an interview where he addressed the question about what the bond market price action is signaling.
The bond market is saying a couple of things to me.
One that inflation expectations are drifting lower - they have drifted lower, and that’s in large part because of the Fed. The Fed is sending these hawkish signals by raising interest rates in a low inflation environment.
And second, I think the markets are pricing in a lower neutral real interest rate. So the interest rate that balances savings and investment in the economy, which is set by broader macro economic forces, has been trending down over the last few decades. I think markets are embracing that concept, and pricing in a lower R* which then caps where bond yields are, and at the same time, can explain some of the appreciation of the equity markets as they are discounting cash flows at a lower rate.
So those are the signals I take away from the bond market right now.
By raising rates in a low inflation environment, we are sending a signal that the 2% inflation target is not a target, but that it is a ceiling and that we are not going to allow inflation to creep above 2% and I think that is putting pressure on the long end of the curve. If you look at the way we have behaved, not what we have said, we say it is a target not a ceiling, over the last 5 or 6 years, we have been treating 2% as a ceiling and I think markets have figured that out, and they are pricing that in.
To me, the Fed is pushing up the front end with their rate increases and pushing down the long end by sending this very hawkish signal about the outlook for inflation.
It’s normal to expect some flattening when the Fed is in a tightening cycle, the front end tends to move up more than the long end does, I would have expected, with the tax reform package coming together, some movement in the long end. But the fact that the long end is not moving either in response to our hikes, or the tax package, that tells me that inflation expectations must be solidly anchored.
Not only does Neel agree with the speculators’ view that the Fed is causing a flattening of the yield curve, but he wants to actively resist it. So far, Neel’s view is in the minority at the Fed, but there can be no denying that some FOMC board members are becoming more sympathetic to his argument (including Yellen - Everything’s Great).
All of these conflicting influences make yield curve trading difficult.
Yield curve versus previous yield peaks
But let’s take it back to 13d’s more important chart in the world. What did the yield curve do during those other trendline peaks?
As you will notice, those peaks often coincided with a bottoming in the yield curve. The trouble is, those bottoms occurred when the yield curve was inverted, or at least at zero.
But today, we are still 55 basis points away on the 2-10 year spread.
So does that mean this isn’t the peak in the 10-year yield?
I am not sure, but I do know we need to take into account the possibility that Central Bank quantitative easing has distorted the yield curve to the point where it no longer functions properly.
Have a look at the Japanese JGB 5-20 spread over the past three decades.
Apart from the initial aftermath of the 1989 Nikkei stock market collapse, the yield curve has since not inverted.
Does this represent the new reality of post super credit cycle collapse? Do the old rules about yield curve inversions need to be thrown out of the window?
The million dollar questions
To answer this question, I think you need to forecast two different economic factors.
The first is whether the recent rise in rates will be enough to derail the economy. If you are a big believer in the 4 Ds (debt, demographics, deflation = destiny), then you should be buying 30-years by the fistful as each successive yield peak will be lower, and 13d’s trendline kiss will most likely be enough to resume the trend toward ever lower long term interest rates.
However, if you are in my camp, and believe that monetary and fiscal response will ultimately determine where interest rates are headed (as opposed to some sort of preordained destiny), you need to ask yourself an important second question. What is the Fed’s (and other Central Banks’) reaction function? And how does the recent change in fiscal policy affect this forecast?
So far, I have been flat out dead wrong on my prediction that the yield curve will steepen. It’s been a terrible call.
I have been wrong for two reasons. The first is that I expected the Federal Reserve to adopt a more Kashkarian approach.
I thought the Fed would lag raising rates. I did not expect them to get out ahead of the curve. Wait! I know that many market participants believe the Fed is way behind the curve in their rate rises, but if that was the case, why has the curve flattened and the US dollar risen over the past couple of years? Regardless of whether you believe a “let-the-economy-run-hot” policy is right or wrong, I expected the FOMC to adopt a much easier policy. This was not the case.
The second mistake was assuming the ECB and BoJ’s extreme monetary policy would not leak into the US. For Quantitative Easing to be effective, monetary stimulation has to be met with private sector credit creation. Unfortunately, in both Europe and Japan, the private sector response has been muted, and instead of creating inflation in their respective economies, the stimulus just ends up leaking out to other markets. This has driven the US long end lower in yield than would otherwise be the case.
But if you think about the coming quarters, both of these influences will be negated. The Federal Reserve appears to have taken the approach that hitting their 2% inflation target is all that matters. With inflation sagging, they are guiding towards rate rises at a more measured rate. In addition, both the ECB and the BoJ are giving signals that the extreme monetary stimulus will be reduced in the coming quarters.
When you combine those factors with the US tax cut which will be both economically stimulative, and also need funding, we have the potential for a real steepening of the curve.
How to trade it
Regardless of whether you are in the flattener or steepener camp, I am going to lay out exactly how you go about trading the yield curve using futures. Let’s start with my favourite - the 5-30 year spread.
The 5-30 year spread has collapsed from 250 bps in 2013, all the way down to less than 60 bps today.
This is knife catching at its best (worst), but here is how you put on the steepener. The two contracts you need to trade are the FVH8 and the WNH8 (ultra long bond) contracts. The ultra long bond contract also has a root of UB in certain trading environments.
To calculate the proper ratio, Bloomberg uses the cheapest-to-deliver bond, and then balances the two contracts using the bpv (basis point value). Plugging it all in, you get the following:
That means for the position to be balanced, a trader needs to buy 100 FVH8 and short 17 WNH8.
Now that’s quite a position, so for us odd-lotters, a more appropriate ratio might be 6-1.
At this level, every basis point move in the 5-30 spread should equal almost $300. So a move from 55 basis points to 60 bps should equate to a profit of approximately $1500 per spread unit. It won’t be exact because the cheapest to deliver ultra-long bond duration is not perfectly 30 years, but that should be close.
To execute this spread, you can just bang out both positions, hitting the bid and lifting the offer, or you can be a little more sophisticated, and trade the spread contract that the CME lists specifically for this position. This contract is priced off the net change from the previous day’s settlement, and will automatically make you long 6 FVH8 and short 1 WNH8 (UBH8).
There, now you can trade like a local.
And if you want the contract ratio for the 2-10 spread, here you go:
Although I didn’t get my curve steepening trade right, one area of the curve that I have been lucky enough to have made the correct call is swap spreads.
When the 2008 Great Financial Crisis hit, banks withdrew their balance sheet, and much to everyone’s surprise, swap spreads went negative. This should theoretically never happen as investors should never be willing to accept a lower yield from a contract with a bank versus risk free US treasuries, but not only did swaps dip below zero, but went deeply negative.
However, over the past year I have argued that Trump’s deregulation push meant banks have returned to extending their balance sheet, and when combined with a few other technical reasons, swap spreads would return to more “normal” levels.
So my 5-30 steepener has actually been long five-year futures, and short 30 year swaps. This spread is only down 35 basis points this year, as opposed to the traditional 5-30 treasury spread declining 65 bps.
And for those that want to trade this spread in the futures market, here is the position hedging calculation.
This trade of long five-year futures versus short 30-year swap futures is still my favoured way of playing for a steepener in the yield curve. Although I understand the arguments as to why the Fed might continue to flatten, and why 13d’s world’s most important chart might indicate the long end might be set up to rally (and further flatten the curve), I suspect that investors are overestimating the Fed’s willingness to flatten the yield curve, and underestimating the amount of stimulus that has been sitting fallow in the financial system, but yet might just have been ignited over the past year.
I still think, the next major crisis will be caused by Central Banks losing control of the bond market. And although it is a lonely trade, I think buying steepeners is still a great trade down here.
At the very least, I hope my post has helped some of you figure out how to bet against me if you don’t agree.
Thanks for reading,