Every year for the past four, Goldman started off the year with an undauntedly optimism and a bullish forecast, one which usually involved a material increase to GDP expectations and, just as importantly, rising 10Y bond yields. And every year for the past four, it took Goldman a few months before it was forced to trim both its GDP forecast and cut its expectations where the 10Y would end the year.
Moments ago 2016 became the latest year in which Goldman was forced to admit it had been too bullish if not on economic growth (that too will come) then certainly on inflation expectations, and as the bank's Francesco Garzarelli admitted moments ago, "we are lowering our bond yield forecasts in the major advanced economies by an average 30-40bp across the forecast horizon. Specifically, we now see 10-yr US Treasuries ending 2016 at 2.40% and 2017 at 2.75%, from 2.75% and 3.30%, previously. The corresponding new forecasts for German Bunds are 0.50% and 1.00% (compared to 0.60% and 1.00% previously), and those for JGBs are 0.10% and 0.30% (from 0.30% and 0.60% before). Exhibit 13 at the end of this document summarizes the forecast changes."
Why the cut? Because after the bank was finally forced to throw in the towel on its wrong 3 rate hike call last week, it no longer has a catalyst to push a strong inflation agenda. Here's Goldman:
Our new projections reflect (i) a downgrade in the profile for Fed Funds rates announced by our US team last Friday (2 further hikes in the remainder of this year, followed by a further 3 next year, compared with 3 and 4 previously); and (ii) the ongoing absorption of duration risk by the ECB and particularly by the BoJ, delivered in conjunction with negative policy rates.
In other words, much slower growth than Goldman had originally expected, coupled with more central bank intervention and frontrunning of bond purchases, coupled with yield differentials between Europe and Japan where the central banks are actively soaking up all available Treasuries, and the US where for the time being there is no QE.
The forecasts conservatively assume that the current deviation from our Bond Sudoku valuation framework (between 1.5 and 2.0 standard deviations from 'fair') will be slowly corrected over the forecast horizon to one standard deviation over the next 6-9-months and close to half a standard deviation by end 2018. We reiterate our view that yield levels below 1.75% in 10-yr US Treasuries (a two standard deviation event) are unlikely to be sustained unless the macro outlook deteriorates materially.
Translation: expect the 10Y to drop below 1.75% on very short notice.
Of course, Goldman does not want to admit that it is wrong (as in the case of its EURUSD parity call), but rather that the market is, well, broken, and provides the following chart to explain why that is the case:
US Treasuries Are Close to 2 Standard Deviations Expensive Relative to Their Fair Value
Well, if they are so "expensive" maybe central banks should stop buying them and further distorting the market? Just a thought.
Ironically, even Goldman now admits that central banks have broken the bond market:
The policy stance of the ECB and the BoJ is shifting away from cuts in policy rates into deeper negative territory towards 'credit easing' measures, which should revive final domestic demand (rather than help export-oriented sectors via a lower exchange rate). This shift is important because our analysis suggests that the combination of negative short rates and outright purchases of long-term debt put in place by these two central banks have largely contributed to depressing the bond premium globally. If official rates are close to their effective lower bound, and asset purchases are spread over a broader pool of assets, financial conditions are likely to ease and this could lead to a rebuild in inflation premium.
JGBs Have Pulled Down Global Long-Term Rates
And then the following even more surprising admission:
Last but not least, the starting point for bond valuations is extremely stretched, in our view. In the US, our survey-based measure of the bond premium and the term premium estimates produced by the New York Fed are now back to the lowest level since the European financial crisis. Our global measure of the bond premium is also very depressed. We attribute this largely to the distortions in long-dated yields stemming from the ECB and BoJ QEs, and their spillover effects.
Premium on US Treasuries Lower than During the GFC
But the most surprising comment from Goldman is the following warning, which is the first time we have seen Goldman do:
On the policy front, all three major central banks can create financial turbulence if not careful in managing investors expectations. The Fed is tightening with very few hikes priced - a historical anomaly - increasing the odds of a 'rate shock'. The ECB and the BoJ are distorting the price of duration (and in Europe, sovereign credit) through their asset purchase programs. Any unanticipated shift in their behaviour could have magnified effects on asset prices. Consider that in 6-months time, there will be only three left to go until the end of ECB QE. We think that purchases will be extended for longer, but this will be a passage that the market will need to navigate.
And so here we stand, 7 years after "the end of the Great recession", and Goldman has just given its most dire warning that if central banks withdraw their support, or even fail to properly manage investor expectations, they could crush the already broken bond market, leading to a "rate shock", one which would result in "financial turbulence" and have dramatic consequences on all asset prices.
In other words, everything is fine.
Ah, the joys of central-planning and micromanaging the world by a few academics locked inside marble buildings.