Last week the bond market continued to be put to the test, and if it could go wrong, it did go wrong. Across the board stronger than expected economic data, including Friday’s better than expected employment data, provided plenty off ammo for risk behavior that saw the yen make new lows against the dollar, pushing equities to new highs with the Dow
(INDEXDJX:.DJI) making a new all-time high seemingly every day. The upward momentum was not lost on the mainstream media as on Thursday I heard CBS Evening News anchor Scott Pelly refer to the announcement of a new high as sounding like “a broken record.”
The 10-year yield, up 20bps on the week, closed at 2.04% near the highs of this leg with the US bond futures contract completely reversing the recent rally, slicing through my pivot level now 141-16 adjusted for the June contract settling right at 141-00 even. This level has acted as previous support throughout the past month of consolidation, but clearly some technical damage has been done.
Nevertheless despite some pretty severe pressure from risk markets the long end of the curve did not break down and the US bond futures contract still remains within the rising regression channel I have been monitoring. Over the next couple of weeks I expect this level to be tested which rises into the 139-16 level, providing the bond market with a critical make-or-break level to hold.
US Bond Futures Weekly Chart
Equity market investors were very proud of themselves this past week and there were a lot of “I told you so” comments from born-again bullish traders and pundits. However I think these people are missing a very important point. In my experience most equity investors are extremely ignorant of bond market dynamics and oblivious to what drives interest rates or how much more money is flowing in and out of that market. The stock market is a sideshow compared to the bond market. If equity investors think that stocks can rise if the bond market is topping, I believe they are in for a rude awakening.
(See also: Where Have We Seen This Market Before?
I was ridiculed on Twitter this week for suggesting that this market was potentially mirroring 1987, which saw rapidly rising bond yields crash the stock market. One person commented that the equity risk premium was too cheap today relative to 1987. Another said it was silly because interest rates in 1987 were 10%.
I believe these two metrics are shortsighted and highly flawed. The equity risk premium (earnings yields – 10-year yield) when measured against negative real interest rates is a faulty valuation tool. Of course it’s going to look cheap. If the S&P
(INDEXSP:.INX) were 100x earnings the ERP would still look relatively cheap against a negative yield. In terms of the 10-year yield comparison, it’s not the outright nominal yield that matters, it’s the relative interest rate and degree of the move that are more important.
S&P 500 vs. 10-Year With 1987 Overlay
Looking at the correlation it doesn’t look so silly to me. Off the 1.50% 10-year low to Friday’s close of 2.04% you have virtually the exact move on a percentage basis as you did in 1987 when yields rose from 7.0% to 10.0%. When interest rates rise into decelerating nominal GDP, “accidents” can happen. If the 10-year is topping and we get a swift move toward 2.5-3.0% as nominal GDP is decelerating then the stock market that everyone loves to own could suddenly become very risky. This also happened in 2007 when the 10-year took out 5.0% as nominal GDP was decelerating below that level. At that time you heard the pundits and strategists rejoicing that the bond market was discounting a breakout in growth. We all know how it worked out.
(See also: 12 Cognitive Biases That Endanger Investors
It’s important to analyze a rapid rise in interest rates in the context of a market that is extremely leveraged. Not to mention, as I’ve been writing about ad nauseam, today it represents a significant percentage of commercial bank assets which are also extremely leveraged. Friday’s FRB H.8
report showed $2.7 trillion in securities on bank balance sheets representing 27% of total credit assets. Using the 10-year as a proxy, if the current 2.0% coupon went to a 3% yield, you're talking about an 8.0% reduction in price which will trap a lot of capital on bank balance sheets.
Ironically this week’s much publicized Fed bank stress test of an “adverse scenario” that most all passed with flying colors doesn’t even stress for rising interest rates. Most banks’ internal interest rate shock models stress for parallel
shifts of +100 to +300 bps but not a face-melting bear steepener. With record low coupons of negatively convex assets a rapid bear steepener would wreak havoc in the credit markets and on bank balance sheets.
Perhaps the Fed isn’t stressing an adverse scenario for rising interest rates because they believe they are successfully holding them down. Would you believe me if I told you that today the Fed owns no more US Treasury securities as a percentage of the outstanding stock than it has on average over the past 50 years?
Fed UST Holdings Percent of Stock
The Fed currently owns 14% of the outstanding US Treasury market stock, and while that number has modestly adjusted throughout the years, since 1963 the average is 14.8%. In the 10 years preceding the financial crisis the average was 15%, and as you can see on the chart, between 2004 and 2007 they owned over 15% of the Treasury stock -- while they were raising interest rates, no less. It’s not the Fed’s balance sheet that has ballooned; it’s the US Treasury’s balance sheet that has.
By launching QE the Fed is merely playing catch-up to the historical average. They have to know this, so we can only deduce that they have used QE as an excuse to con the markets into thinking they are stimulating asset prices and aggregate demand. In reality they are just maintaining the status quo. It would seem QE is just a PR stunt to engineer inflation expectations and animal spirits. As far as I can tell there is absolutely zero correlation between the amount of Treasuries the Fed owns and the level of interest rates. For instance, in 1970 they owned 21% of the stock and the 10-year yield was at 6.50%, but in 1980 they owned 16% and the yield was 12% while in 1990 they owned only 9.5% but the yield was 8.0%.
Why should we assume that the Fed has been successful in lowering interest rates, or that they would be able to hold them down if the market were ready to turn? Buying $5 billion/day in a market that trades $500 billion isn’t going to cut it. The fact of the matter is that bond yields are low because the economy is weak and there is no demand for money. When interest rates rise in an environment where economic growth is decelerating, bad things can happen -- just like in 1987.
I am not some crash-calling perma bear like many others who have been getting their brains beat in by this rally. In fact I was looking for new all-time highs and a melt-up back during the summer of last year.
On June 12 in Trading the Wrong Playbook Bubble
I pointed to the significance of holding the historic 1265 pivot level while many in the speculative community were positioned for another crash:
I had this pivot area not only because it represents last year's double bottom prior to the August crash, but it also represents the low after Bear Stearns imploded in March 2008. If this market truly is just back-filling the gaps left behind by the financial crisis then making this area of support is very bullish and could point to new all-time highs going into the fall.
Then just over a month later on July 30 in Bernanke’s Astonishingly Good Idea
I again pointed to the negative sentiment and short base as a catalyst for a melt up rally:
Last year the speculative community was still long the QE II reflation correlation trade thinking they were going to get an extension and when they didn’t it was Katy bar the door. This year they are short. If we don’t crash soon when the boys come back from the beach they may be piling in to get long before year end. It could be melt up city.
ES COT Large Specs Net Position
Looking at the nature of the positing vs. the price, it’s pretty clear to me this is what has happened. What could be different about their assessment of fundamental conditions between then and now? The fact is this rally has not been on the back of improving fundamentals; it’s been on the back of a massive short squeeze in the speculative community. With the short base cleared and newfound bullish sentiment suggesting extreme confirmation bias I am beginning to think the risk has overwhelmingly shifted to the downside.
This is not something I take lightly, and I invoke 1987 in all seriousness. The very fact that it is considered “silly” is exactly why it can happen. Markets don’t crash because everyone is expecting it; they crash when people think they can’t go down.
In 2006, as I was trying to short the market, a wise trader and eventual mentor said to me: There will be no bears left when this market tops
. This was invaluable advice and it has kept me on the right side of this rally for a long time, but I am starting to see the sentiment turn. With each little poke to new highs, more and more bears are dropping like flies; body bags are indeed piling up high along capitulation row. After looking to get long at 1265 in anticipation of an explosive rally to new highs I am now looking to get defensive as I think this rally is in the process of completing what will prove to be a cyclical rally in an ongoing secular bear market.
S&P 500 Weekly Count
I want to now take a look at my working Elliot Wave count of the S&P 500, which has served me well since the 2009 lows and should be considered a potential scenario. The count has been evolving and the main thing I want to emphasize is where the count changed into what I believe no one sees. The basic premise and most important objective in counting markets is determining whether you are impulsing or correcting. Generally markets impulse with the trend in five waves and correct against the trend in three waves.
I initially had the first leg of the rally out of the 2009 hole as an impulse wave 1 and the subsequent correction as a corrective 2 wave. However as we rallied out of the 2011 hole the market should have been embarking on the powerful wave 3 up. The first move satisfied the impulse, but as the rally unfolded it took on a more grinding nature exhibiting 3s up rather than explosive 5s. I also concluded that the 2011 leg up wasn’t a 5 wave impulse but rather a Fibonacci .618 extension b wave in a larger B. I changed my 2 to the extremely rare “running flat” B. Realizing the ensuing rally would be a C I started mapping a diagonal.
Thus far the rally has unfolded in a near perfect diagonal and this week we finally got the needed e wave “throw over” of the upper end of the rising wedge. You can see my “e” label was a bit early, but what I have been looking at may have happened last week. That the "throw over" is occurring as the weekly RSI has finally made it to the ceiling is further evidence that the rally is in the late stages.
If I am correct we are very close to finishing a 3 wave ABC corrective move off the 2009 low. This has major market implications. With in the larger trend it’s quite possible we could have finished a larger B wave up with the 2009 low completing an A wave. The ensuing C wave move lower could be a spectacular collapse. You might even call it a crash.