Thursday's "better-than-expected" Q2 GDP report was supposedly cheered by equity investors who suddenly forgot about the emerging disaster in Syria that was meant to be the catalyst for weakness on Wednesday. The number that matters, year-over-year (YoY) nominal GDP, registered a 3.1% gain, up from the previous 2.9% estimate and equal to the 3.1% Q1 growth rate, which was the slowest quarterly growth rate since emerging from the recession in 2010.
The more important data point from the week was Friday's income and spending report, which provides the YoY growth rate of personal consumption, the biggest component of GDP and the first look at the Q3 run rate. The July YoY growth rate of personal consumption (PCE) registered a 3.1% pace after averaging 3.0% in Q2, 3.3% in Q1, and 3.7% in Q4. This key component of US economic growth remains near the lowest levels in the post-WW II era and also suggests the economy remains mired in 3% nominal growth.
With the Federal Reserve's monetary stimulus operating at full throttle since January, it's hard to believe that the economy can only garner a 3% nominal growth rate. As the Fed prepares to dial it back at its September meeting -- known as the so-called "tapering" of QE -- it will be interesting to see if we can even continue this modest pace. Stimulus or not, it is likely that this 3% growth rate represents the new structural growth potential of the US economy.
Based on the latest FOMC
forecast, the Fed is projecting anywhere from 50 bps rise to a 3% Fed funds (FF) rate by year-end 2015, with a large consensus projecting a 4% rate in the long term. The Fed probably gets this number based on what it perceives to be the potential growth rate of nominal gross domestic product (NGDP), which averaged 5.5% through the 1990s and 5.1% from 2000 up until the financial crisis. However, on the back of the most stimulative Fed in modern history, growth is settling in at only a 3% rate. Can a 3% growth rate handle a 4% overnight lending rate?
Over the past 40 years, the FF rate has not spent a lot of time above the growth rate of NGDP without coinciding with (or causing) a recession, with the lone exception being the tightening cycle in 1994. In fact, the last time the FF rate was at 4% was in 2005, which arguably began the process of popping the credit bubble. At the time, 4% was still far below the 6.5% growth rate in NGDP. In fact, if you look at the average spread of NGDP growth to the FF rate going back 40 years, it is 70 bps, which at the current 3.1% NGDP growth rate, projects only a 2.4% FF. Odds are, if the Fed hike takes the interest rate to 4.0% in the context of 3.0% growth, money would get very tight, the curve would invert, and a recession would follow.
Nominal GDP Vs. Fed Funds Rate
Similar to the relationship between FF and NGDP, the 10-year yield tends to find equilibrium around the growth rate of personal consumption. Why is this? When the risk-free rate exceeds spending growth such as it did in 1990, 2001, and 2009, due to a collapse in aggregate demand (which resulted in a recession), the 10-year subsequently fell to ease credit conditions. Long-term interest rates will calibrate against the growth rate in nominal spending, lowering the cost of money when spending weakens to stimulate growth, and raising the cost when spending accelerates to tighten.
10-Year Vs. PCE
One of the perils of QE, which is exacerbated by the fact that it has lasted so long, is that the market has not been allowed to calibrate around the growth rate of consumption. The volatility we have thus far been witnessing is The Yield Curve Adjustment Process
I wrote about on May 13.
Whether you believe the Fed has successfully manipulated the yield curve or not, just the fact that it is in the market means that no one really knows where Treasuries should trade. For this reason, we don’t know where asset prices should trade. However, it seems that we are about to find out, and I think it is going to be a very uncomfortable process.
This is exactly what is happening. You have a lot of very levered bond market participants who own a lot of low coupons suddenly trying to figure out what is fair value.
The dealer strategists who blew up their clients' investments by convincing them that the Fed was controlling the long end of the curve are still trying to calibrate their interest rate forecasts based on whether the Fed is going to be buying. This focus on tapering shows a complete ignorance of what drives interest rates and where we are in the cycle. As I have been saying ad nauseam, the amount of bonds the Fed buys is irrelevant. It’s not the flow or the stock; it’s the discount. Many participants misinterpret the impact of QE on bond yields. Ironically, QE lowers nominal yields via raising inflation expectations and thus lowering real yields of which nominal yields have followed. The reverse is also true. The entire back-up in nominal yields has been a function of real yields rising as the inflation premium falls. Tapering should continue that trend as the inflation premium is reduced; ultimately, nominal yields are going to settle in near the discount for growth plus this inflation premium.
In reality, QE became irrelevant about as soon as it got started when the Fed backed off their full commitment to their inflation target. Upon the release of the January FOMC minutes in February, we found out about the Fed’s ex-post QE cost/benefit analysis, which clearly indicated that the committee was no longer comfortable with the side effects of the program. Since then, the long end of the curve has essentially been free to trade, and this entire year has been about finding equilibrium. First, the long end will set the price, and then the Fed will follow suit.
What is the equilibrium Fed funds rate?
Kevin Ferry's Fed Funds Model
Based on Kevin Ferry’s model
(the go-to guy for money market interpretation), the current Fed funds rate at ~10 bps (0.10%) is already neutral
. The benchmark for which he measures this equilibrium is the spread between FF and 12-month LIBOR, which at 66 bps has not even budged since the rest of the Eurodollar curve for 3-month LIBOR futures exploded in May and June. If the market was responding to an increase in the demand for money, these spot LIBOR rates would be increasing with the futures curve. They are not. The Fed thinks that it can go to 4% when at 10 bps, it’s already at equilibrium? I don’t think so.
The Fed may well hike FF to 4%, but it will massively invert the curve. If the Fed even gets to 2.5% by the end of 2016, I think we could see a scenario where the whole curve flattens to zero, converging at 3%, which looks to be the new structural nominal growth rate for the economy. Think it can't happen?
5-Year/10-Year Vs. 10-Year/30-Year
It wasn't that long ago that the 5-year/10-year spread and the 10-year/30-year spread were both flat at 5%. In 2006, when the Fed pushed the Fed funds rate to 5% with nominal GDP decelerating due to the credit bubble unwind, the whole curve went to zero. Similarly, in 2000, when the Fed pushed the FF rate above 6% in line with NGDP, the whole curve inverted. This is not random. This is what happens when the cost of funds exceeds the growth rate.
In this cycle, as the Fed tightens into what I posited
last week was potentially an already-tight monetary condition with decelerating growth, I think curve convergence is going to happen again. As the Fed shifts from ultra-easy policy to neutral, we are going to begin to see the yield curve reflect this uneasy reality by normalizing rates in the front while taking the term premium out of back, flattening the curve.
Last week, with little fanfare, the spread between the 10-year and 30-year flattened to 91 bps, the lowest level since the 2011 stock market crash prompted a massive flight to quality bid in the long end. This is where the inflation premium beta gets the most action, and if the market is reversing the QE steepening bias that has ruled the bond market since 2008, we are in the midst of a cyclical shift.
There is no greater consensus market view right now than "rates are going higher." Everyone wants to be short the long end of the curve because the Fed is set to taper and soon will embark on a tightening cycle. It may happen in the short run, but in the long run, the curve doesn’t work that way. The bond market is not waiting on the Fed to tell it where to go. The long end of the curve is an inflation premium discounting mechanism, the QE inflation discount is in the rearview mirror, and the quest for equilibrium is now governing the term structure.
Where is the curve’s equilibrium?
As I said a couple of weeks ago in Interest Rates Are Normalizing
, interest rates shot straight to their respective 5-year means:
With the bond market open to trade the 10-year, it immediately reverted to the 5-year mean. Not only did the 10-year revert to the mean, but the 5-year went to 1.60% versus a 5-year mean of 1.65%, and the 30-year went to 3.75% versus a 5-year mean of 3.75%.
By my math, the long end of the curve is pretty close to fair value, and arguably, the 30-year at 3.75% is cheap relative to what could be a much slower structural growth rate of NGDP. The initial move of the tightening cycle was a normalization of the long end. Once the Fed begins to actually tighten, we shall see a normalization in the short end.
In my view, it’s a 3% world we live in. The scenario I see playing out is one where growth rates remain at 3% the FOMC raises FF to 3% over the next three years, and the 5-year, 10-year, and 30-year all converge toward 3% as the bond market discounts tight monetary conditions and virtually zero inflation premium. That’s because... "Three is a magic number.
Yes it is, it’s a magic number."