By Jeffrey P. Snider
In January 2011, FRBNY began surveying primary dealers in order to try to gain a better understanding of how normalizing policy after two large (in their terms) programs of quantitative easing might affect money and bond markets. The banks were asked a series of questions, an array that has evolved over time, mostly about expectations for the future track of Federal Funds or some economic account. The Fed branch would also ask performance questions, such as respondents' views of communication effectiveness on the part of the FOMC. Principally, it seems, the central bank was expecting to see if any irregularities could be anticipated as the world began to understand and incorporate the eventual policy exit.
It didn't quite happen as they expected, of course, as just a few months later normalization was the last thing anyone was figuring. The surveys kept going, however, becoming somewhat more prominent and relevant toward the end of QE3. The December 2013 survey, just as taper had become reality, asked dealers to, among other indications, forecast the benchmark 10-year US Treasury rate in light of the change in monetary policy. About 70% figured that just one year later, by the end of 2014, the 10-year CMT yield would be something greater than 3%.
That seemed to be the consensus, particularly as nominal rates rose from the middle to the end of that year, with the 10s actually touching 3% if only briefly to that point. It has been a sport, of sorts, to claim the end to the "bond bubble." This is not to say that there isn't a bubble or at the very least bond valuations imply something other than investment considerations. Yields are relative, though, and relativity doesn't necessarily mean what everyone seems to think.
In the middle of 2013, as the bond market, particularly MBS, was in full disarray, claiming the end to the supposed 30-year bond cycle was easy. From June 21, 2013, coincidentally the next to last day of the MBS selloff:
Investors in bonds are being warned that the much heralded bursting of the bubble is about to happen. Last summer, one leading bond manager broke ranks to say that he would not buy his own asset class. His reason? Within a year the bond bubble would burst. This week the same manager is saying that time is upon us.
Not even junk bonds were all that much affected in 2013, as yields kept piling lower and lower beyond even taper. It wouldn't be until the dramatic change in risk, credit risk not rate risk, in the middle of 2014 that junk bonds would finally reverse. That led to this "unexpected" dichotomy ultimately leaving the dealer surveys in tatters. The 10-year yield would start 2015 below 2%, not above 3%. There is a world of difference in between those levels.
Survey responses in the subsequent years would not get any better, though they do contain an important clue. At the end of 2014, more than half of dealers expected the 10-year to be less than 2.50% at year-end 2015, but that proportion would drop to just 15% by the December 2015 survey (which was distributed on December 3, 2015, and received back at FRBNY on December 7). What changed was, of course, that by early December 2015 everyone expected the FOMC to increase the Federal Funds rate, which they did. What nobody apparently accounted for was that the bond market didn't care at all about FOMC direction or interpretation.
By January 1, 2016, the 10-year yield was right back near 2% and heading lower again; meaning that respondents a year out in 2014 were much closer to correct than only days away at December 2015. In short, predominant views of monetary policy made dealers more wrong (for lack of better phrasing) even though they had every advantage of time. This gets to the heart of the world's bond problem, or at least how the mainstream keeps figuring it all wrong.
Based on that new model and statistical norms, there's less than a 1 percent chance U.S. 10-year yields fall below 1.1 percent, especially as the Federal Reserve moves to raise interest rates.
But here's the punchline:
"We had to revert to a model-based approach to figure out how low yields can go after we broke below 1.4 percent," a scenario that the firm didn't think would happen unless the Fed did an about-face, said Subadra Rajappa, SocGen's head of U.S. rates strategy. [emphasis added]
This model now pegs the "fair value" of USTs 10s at 1.95%, though as Bloomberg points out, rightly, it figured "fair value" at 2.85% before this latest reconfiguration. These mainstream models and interpretations are all GIGO - garbage in, garbage out. Nobody can figure out bonds because nobody can figure out money, and thus what real monetary policy is. Yet, the answer is so obvious that no one seems to want to believe it. As that article I quoted above from June 2013 reads, everyone is wrong based on one, simple miscalculation.
The economist predicted that eventually, thanks to economic recovery, these bond yields would rise significantly, inflicting large capital losses on their holders. That's because as yields rise, prices fall.
They all assume, all have assumed, that economic weakness is just some temporary phase, an inconvenient speed-bump that will surely fade as monetary genius further asserts itself. It was a dubious prospect in 2013 where unbiased analysis of the global economy only showed increasing doubt and trouble, but it is much more so in 2016 where "transitory" can no longer apply.
In other words, the bond market signals only further and further economic difficulties no matter what the Fed says and does, but these bank teams and economists don't believe the market, choosing to instead cling to orthodox economic interpretations to the point they have to rethink their models just so they can reach the same conclusions that were wrong before. They ignore the economy as it is, described all-too-well by bond and funding markets, in favor of an economy that "should be" as described in theory and academic mathematics. Reality is virtual for them, a computer simulation more "real" than bond prices; belief and faith in central banks absolute no matter how many years already proved to the contrary.
These are mistakes we have seen before, primarily during the Great Depression. Economists and policymakers then were also more confident in themselves despite the contradiction of the bond market, especially UST rates. No matter how much gold entered the US after the January 1934 devaluation (default), treasury yields only declined (though quite gently, but steadily).
This was and is now the interest rate fallacy. The amount of gold throughout the 1930s was misleading, as it didn't suggest anything other than what official holdings tallied to. It was only assumed that the increase in gold meant an increase in money supply (and circulation) in the real economy. Despite the fact that the economy did appear to recover, and that consumer and producer prices rebounded often sharply, they never did reach their prior peak, leaving the economy essentially shrunken after the Great Collapse (which is what we also find today after the Great Recession).
Gold stock, or even "high powered money" as Milton Friedman measured, did not translate into actual monetary growth - a fact that the UST yields demonstrated. Low and lower rates were signaling instead "tight" money not as a condition of policy or in official holdings, but as the more important matter of functional money (unseen) within the real economy.
The 10-year yield began to decline long before gold began pouring across the US border after FDR's Executive Order revalued the dollar. The decade-long decline in rates began at the outset of 1932, two years before the default event. It takes but one additional variable to more easily infer the cause of real money contraction and thus the steady trend of malfunction exhibited by bonds from that point on.
The first wave of bank failures toward the end of 1930 was bad and ominous, but the second that crested around October 1931 was systemically altering. The effects were not strictly monetary, of course, as it "encouraged" a paradigm shift in behavior that no monetary policy or fiscal dictation could overcome. In raw and general terms, true, unmeasurable monetary circulation was just never the same - until two decades later, long after WWII into the 1950s.
I think we have seen something very similar in our own experience of the Great Recession. The first shock (which was really two waves, the first in August 2007 through Bear Stearns in March 2008; the second starting July 2008 through the end of the panic in March 2009) was likewise bad and formulated into a great crash. The second, like the second wave of bank failures in 1931, was the truly fatal blow - the crisis in 2011 that shows up in every economic account and a great deal of what we can measure of the eurodollar system.
And so the mainstream keeps calling for a bond market reversal based on a fundamentally flawed proposition; a recurring mistake in just these kinds of extra-cyclical conditions where the usual rules and expectations of "stimulus" and policy just don't apply. The UST market, among other bond classes, as it goes only lower (in yield) tells us that the economy is getting worse (which is what we actually find here and all over the world). Economists and econometrics don't believe it because they are hardwired not to; to them, monetary policy always works, therefore rates will have to rise as the economy that should be must be just around the corner.
From that utterly backwards view, persistently falling interest rates are an absolute mystery; a riddle without answer. The proper, unbiased perspective shows there is no riddle at all, only that the bond market (and "dollar" markets) has been right this whole time. UST yields are not backward-looking proof of academic concepts, they are a tested and verified forward-looking warning of what is to come. Sometimes, as in the 1930s, what is to come stretches on interminably, long past what used to make sense.