Continuing the recent trend of models are right, reality is wrong economic "analysis", launched when some unknown goalseekers over the past month suggested that a double seasonal adjustment is appropriate to any data that does not conform with the mandate of growth at all costs, and which has since been proposed as official Fed policy suggesting it is only a matter of time before the BEA rewrites history and drastically "adjusts" historical GDP numbers significantly higher [7], overnight Goldman was the latest to suggest that contrary to popular opinion of pervasive economic weakness, what the Fed sees in the economy is actually "little slack."
Specifically, Goldman updated the Fed's infamous FRB/US computer model [8]which estimates key components of the economy's supply side, including potential output and the structural rate of unemployment, and has the following take-aways:
- potential growth has continued to recover (estimated at 1.8% year-over-year in 2015Q1);
- the structural unemployment rate has remained low at 5%;
- the cyclical gap in participation is very small; and
- the output gap has continued to close (reaching -0.6% in 2015Q1).
In other words, even without a double seasonal adjustment, the Fed may very well surprise with not only a September, but even a June hike. After all recall that to Yellen stocks are now clearly overvalued, and the cornered Fed Chairwoman is between a rock and a hard place - keep failing to rase rates and risk another bond tantrum as all the shorts are squeezed leading to even more illiquidity and volatility, or slowly take the air out of the stock bubble (good luck with that).
Some more details on Goldman's simulation of FRB/US:
In 2014, Fed staff economists published a model that estimates key components of the economy's supply side, including potential output and the structural rate of unemployment (the details are available here). The model employs statistical "filtering" techniques to separate trend from cycle and the resulting estimates form the supply side block in the Fed staff's FRB/US model. One of the key features of the model is that it uses not only GDP to estimate the cycle and trend in activity but also other information including gross domestic income and labor market data. Using a range of indicators to measure growth is a desirable approach in our view and one that we have used to construct our current activity indicator (CAI).
So in order to gauge what the Fed is really thinking based on its own computer feedback, overnight Goldman updated the Fed staff's supply side model through 2015Q1.
"Most of the required data for the model is already available for 2015Q1, including GDP, labor market and price information. For variables that are not yet available for 2015Q1--including, for example, gross domestic income and a couple of custom-made series--we use either the closest available proxy or the Fed staff's latest assumption (available in the March 30, 2015, update of the FRB/US model). We also project the model forward using the latest Fed staff's assumptions for the exogenous variables of the model (such as population growth, labor quality and the evolution of capital services)."
The model's findings:
- First, potential growth has continued to recover after slowing notably during the crisis. The estimates suggest that potential GDP growth slowed from almost 3% before the crisis to only 1% in 2011. Since then, estimated potential growth has recovered, reaching 1.8% year-over-year in 2015Q1. The intuition for why potential growth has continued to recover despite very weak GDP growth in 2015Q1 is that the model also puts weight on growth in employment (which has been solid) and gross domestic income (which is expected to be solid in Q1). Looking ahead the model suggests that potential growth will rise a bit further to 2% over the next couple of years, due primarily to an assumption that capital services continue to normalize.
- Second, the estimated structural unemployment rate rose in the aftermath of the crisis to a peak of 5.8% in 2010 but then declined to 5% in mid-2014. The model suggests that the structural unemployment rate has moved sideways since then and projects that it will remain at 5% going forward.
- Third, the model explains the vast majority of the decline in the labor force participation rate since the onset of the crisis by structural forces. Of the observed 2.8 percentage points (pp) decline in participation, the estimates attribute 2.6pp to structural factors and only 0.2pp to cyclical factors. Looking ahead the model projects that the trend participation rate will continue to decline, albeit at a slower pace, reaching 62% in 2020.
- Fourth, the estimates suggest that the output gap has closed rapidly since reaching a peak of almost -8% in mid-2009, reaching -0.6% in 2015Q1. The closing of the output gap slowed in 2015Q1 but continued due to solid growth in activity measures other than GDP (such as employment).
The bolded is key, as it is the closest to explaining why the Fed is in such a seeming rush to find justifications for hiking. Visually:
[9]
Does Goldman, which anticipates a September rate hike, agree with the Fed's computer simulation of the US economy? Not entirely:
The most important reason why we continue to see more slack in the participation rate is that wage growth--historically one of the best indicators for measuring labor market slack--remains sluggish. A likely reason why the Fed staff model points to less slack is that it does not use wage inflation in its estimation of labor market slack.
Said otherwise, it once again comes down to just one thing: wage growth, or the lack thereof. We have extensively shown before why there simply can not be wage growth for the bulk of the US population [10], ironically precisely because of the Fed's erroneous micromanagement of the economy, something which both the Fed and Goldman must realize by now. So if indeed the Fed is driven by this variable, expect no rate hikes for years, and an S&P of 3000, 4000 or whatever the breaking point is for 90% America's population to go "French Revolution."
If, however, the Fed's concerns about a stock bubble are dominant, the Fed will be willing to ignore the lack of wage growth, and proceed with at least one rate hike in the coming months just to test and see if it can even do it.
But perhaps the best hint of what the Fed will do comes not from the Fed, not from FRB/US or from Goldman, but from Gartman, whose overnight note has this:
- In summary then, here will be no rate increase this year. Count on it.
And now we know.
