It was just two days ago when we observed that [12]during the latest Fed matinee, none other than Goldman's Jan Hatzius presented a slide deck suggestively titled "Hiking Rates in the Name of Financial Stability."
Said slide deck, in addition to hinting that the macroeconomy (not Goldman's bonus pool mind you) may not be able to withstand the shock and horror of a 0.25% rate hike, also contained the following binomial decision tree (in which Goldman, with a straight face said that it was unclear if rate hikes "reduce bust risk") designed to strongly "clue" the academic central planners in the Marriner Eccles building just how and when to act.
[13]
It was here that Goldman clearly stated two days ahead of today's non-farm payroll number, that the US economy "may not be ready yet", after nearly a decade of zero interest rates, for a rate hike.
We will ignore the irony that the economic weakness is taking place over a year after the same Jan Hatzius predicted that the US would grow at an "above consensus" rate in 2014 (forecasting 2.9%, the final outcome 2.4%) and further. It took place precisely three months after Goldman predicted a 3.0% GDP growth rate in Q1 2015, a forecast which Goldman since cut to 1.1% (it was under 1% until the February import crash provided a bean-counting gift to the GDP trackers). We will certainly ignore Goldman's forecast of a 2.6% 10Y yield on March 31 just three months prior.
[14]
What we won't, however, ignore is a note released several hours after today's disastrous jobs print by the Goldman economic team, titled "It Is Hard to Be Reasonably Confident", in which Goldman takes a machete, for the second time in 4 years [15], to its "above consensus" forecast.
Here is what Goldman had to say:
Today’s employment report was a disappointment, as payrolls posted a weaker-than-expected gain in March and employment gains were revised down in prior months. The FOMC’s labor market condition of “further improvement” nonetheless strikes us as a manageable hurdle. Bad weather likely weighed on the latest employment report, slack measures continued to narrow in March and the unemployment projections in the SEP suggest that the FOMC was already expecting a slowdown in the pace of employment gains in coming months.
The inflation condition is a much higher hurdle. Given recent inflation news—including the deceleration in year-over-year core PCE inflation to 1.4%, nominal wage growth of 2% and declines in some indicators of long-term inflation expectations—being “reasonably confident” in inflation returning to the 2% target is a tall order.
Ok, so the end of the waiter and bartender renaissance is a hurdle, but not nearly as bad as the realization that now the US will have to import deflation not only from Japan but also from the Eurozone and deal with an oil slump that contrary to Goldman's predictions isn't going to dead cat bounce any time soon, and if anything will only get worse.
Then what?
We use the Fed staff’s FRB/US model to gauge the uncertainty around the inflation outlook. Specifically, we draw “shocks” from the historical behavior of the economy since 1997 and trace out the implications for the evolution of inflation in the model.
Oh, so Goldman will use a license of the model that the Fed itself uses to forecast the future. Surely we can expect projection certainty within 0.1%.
Of course, Goldman knows this is ridiculous, so it gently unveils the first snowman: telling Yellen how Goldman would think... if it were Yellen (not because the old academic is utterly clueless what to do at this point, mind you - just our of courtesy).
Chair Yellen recently discussed a number of factors that would affect her confidence in the inflation outlook. These include changes in: (1) the pace of labor market improvement, (2) core inflation itself, (3) wage inflation, and (4) longer-term inflation expectations.
We next explore the effect of changes in these economic conditions on our confidence metric in FRB/US. We start with an example of how shifts in long-term inflation expectations might affect the uncertainty around the inflation outlook. Specifically, we trace out the distribution of inflation in 2017Q4 in response to a ¼pp increase and decrease in long-term professional forecaster inflation expectations in 2015Q2. Exhibit 3 shows our results. We find that the confidence score drops to 51% when inflation expectations tick down and rises to 68% when inflation expectations move up.
[16]
So kind of Goldman, to do all in its power to try to boost the Chairmanwoman's confidence.
Actually, we misspoke, because next up Goldman finds that contrary to its perpetual bullishness of the past 5 years, "the bottom line from these scenarios is that plausible improvements in any single indicator are unlikely to deliver reasonable confidence on their own."
Exhibit 4 summarizes our results for a range of indicators cited by Chair Yellen. In addition to the inflation expectations experiment above, we also consider shifts in wage inflation, changes to the growth outlook over the next four quarters and news on the trend in payroll gains (that again last for four quarters). We only consider one shock at a time and trace out the resulting confidence score for a range of values. We report the confidence score for both the end of 2017 (black line) and for the end of 2016 (gray line). The key take-aways are as follows.
First, the simulations confirm that shifts in long-term inflation expectations have a large effect on the distribution of inflation outcomes at either horizon. Second, changes in the pace of wage inflation have a meaningful effect too. For example, a 0.5pp pickup in the pace of wage inflation raises the likelihood that inflation will exceed 1.75% at the end of 2017 from 60% to 64%. Third, shifts in the growth outlook over the next year—which affect not only real GDP growth but also payroll growth—have significant effects on the inflation outlook. For example, we find that a ½pp reduction in the expected growth pace over the next four quarters lowers the confidence that inflation will exceed 1.75% at the end of 2017 from 60% to 51%. Fourth, our simulations suggest that shifts in the pace of payroll growth alone—i.e. productivity shocks—have small effects on the inflation outlook. Payroll disappointments, in other words, are probably not particularly informative about the inflation outlook unless they reflect a broader growth slowdown. The bottom line from these scenarios is that plausible improvements in any single indicator are unlikely to deliver reasonable confidence on their own.
[17]
Goldman leans in some more just before unleashing the roofie bomb, because having run out of credibility, the vampire squid now does a "how to become more confident" bridge chart: one assumes they tasked the recent Harvard grad with this one. Remember: when the economy is growing at 0.1%, one's best bet is to just chart how one gets from "unconfident" to "confident."
Simultaneous shocks are more likely to do the trick. Exhibit 5 provides an example of what could raise the confidence score above 75%, a likelihood one might consider to qualify as “reasonable confidence.” The chart shows the marginal impact on the confidence score of a combination of shocks. A ½pp improvement in the growth outlook would raise the confidence score by 8bp, a ½pp pick-up in wage growth on top of that would raise the confidence score by another 4pp and an additional ¼pp shock to inflation expectations would be enough to boost the confidence score for the combined scenario to over 75%.
[18]
It Is Hard to Be Confident
The FRB/US model is a helpful tool for analyzing the uncertainty around the inflation outlook, but it comes with a number of caveats. Inflation expectations, for example, play a very important role in determining inflation 2-3 years ahead, resulting in relatively minor roles for recent news on payrolls or actual inflation. Our analysis is therefore best seen as an illustration of the issues involved.
Good, because if your previous "analysis" is any indication of how accurate the current one will be, it's best to just ignore any "projections" and certainly those 2-3 years ahead.
In fact, it is best to just ignore pretty much everything said to this point which is nothing but econobabble gibberish. Because here, at the very end of the note, is where Goldman finally says what it meant to say all along.
The analysis suggests that it is hard to be “reasonably confident” in the inflation outlook given current economic conditions, unless several inflation drivers rise at the same time. We therefore do not have much confidence in the inflation outlook and believe that the right policy would be to put hikes on hold for now.
And there you have it - from a Q1 GDP forecast of 3% and a June rate hike, then to a September, then to a December, then after shocking "snow in the winter" finally caught up the the BLS' seasonal adjustments Goldman has finally come out of the closet and dropped the first (of many) trial balloon.
Only this is not the trial balloon for the indefinite postponement of rate hikes, those were never going to happen in the first place, but a trial balloon for what the real endgame is here: QE4... just as we most recently predicted would happen 2 weeks ago [19] (and as distantly as mid-2013). Because in a world in which global central banks are already monetizing 100% of global net issuance, there is no longer any need for subtlety: either we desperately try get inflation through unlimited printing of money, the more sharp and acute the inflation spike the better, or the entire fiat system implodes. And as long as central banks have something to say about it (and courtesy of Bernanke's helicopter speech we know precisely what their last action will be), it will be the former.
Finally, what goes without saying, is that in a world in which the Federal Reserve is a branch of Goldman Sachs, what Goldman Sachs trial baloons, Goldman gets.
[20]
