Is it a mutiny? A family feud? Sibling rivalry? Or an attempt to provide more accurate estimates of economic growth?
Greater accuracy would certainly be a good thing, but that may be the least plausible explanation for the New York Federal Reserve Bank’s launch of Nowcast.
Traders need real-time economic data that they can use to guide trading decisions. Providing it has been a role that the Atlanta Federal Reserve Bank took on when it created GDPNow in July 2014. Today, GDPNow is widely used by traders—even though, as we’ve previously noted, Fed projections of economic growth are almost always overly optimistic.
So why add a second source of inaccurate estimates of GDP growth? There are several possible explanations:
- New York is the center of the universe and New York Fed Chair William Dudley found it unacceptable that the Atlanta Fed was grabbing all of the attention with its GDP forecasts.
- With two different forecasts, the Fed doubles its probability of getting it right (although two times zero is still zero).
- The Fed wants to show that it does more than just manipulate the stock market and buy bonds.
- The New York Fed needs to justify its budget and figured it is a good use of taxpayers’ money to produce two separate and distinct GDP growth estimates.
- The New York Fed wants to prove that it is as capable as the Atlanta Fed of producing inaccurate estimates of GDP growth.
Whatever the reason, the second Fed GDP growth estimate was added “to the puzzlement of some traders,” as The Wall Street Journal put it.
“Puzzlement” may be an understatement, given that the Nowcast estimated annualized growth for the first quarter of 2016 to have been 1.1%, while GDPNow estimated growth at 0.1%. Which is accurate? Was growth tepid in the first quarter, as the New York estimate suggests, or was it non-existent, as the Atlanta estimate suggests?
A difference of 1% in the growth estimate is pretty significant; with growth as low as it is, that represents a difference of 1,100%.
We wouldn’t count on either estimate being accurate. Nor would we follow the advice of Torsten Slok, chief international economist at Deutsche Bank AG, who is suggesting that clients take an average of the two Fed numbers until both have longer track records.
Keep in mind that these central bankers are the same people who thought that buying trillions of dollars’ worth of bonds was the best thing that could be done to lift the country out of the 2007 financial crisis, which is why the rate of growth has been abysmal for years.
Many private financial institutions create their own real-time GDP numbers, but, for most of us, it’s enough to know that the economy is continuing the dismal under-performance that started in 2007 and has lasted throughout the Obama administration’s tenure of over-regulation.
Behind the Numbers
But, if you really want to know what’s behind the Fed numbers, consider the methodologies used by the two Fed banks.
On its website, the Atlanta Fed explains that, “The GDPNow forecast is constructed by aggregating statistical model forecasts of 13 subcomponents that comprise GDP. Other private forecasters use similar approaches to ‘nowcast’ GDP growth. However, these forecasts are not updated more than once a month or quarter, are not publicly available, or do not have forecasts of the subcomponents of GDP that add ‘color’ to the top-line number. The Atlanta Fed GDPNow model fills these three voids.”
We’re not sure what “color” the Atlanta Fed is using in its forecasts, but it’s probably rose.
Nowcast, meanwhile, claims that, “Our model produces a ‘nowcast’ of GDP growth, incorporating a wide range of macroeconomic data as it becomes available. With this approach, we aim to read the real-time flow of information and evaluate its effects on current economic conditions. The platform provides a model-based counterpart to the more routine analysis at the bank, which has traditionally been based on expert knowledge.”
So a wide range of macroeconomic data is used to calculate economic growth. Who knew?
“The platform employs Kalman-ýltering techniques and a dynamic factor model,” the Nowcast site explains. “The approach has a number of desirable features. It is based on:
- A reliable big data framework that captures in a parsimonious way the salient features of macroeconomic data dynamics;
- A design that digests the data as ‘news,’ mimicking the way markets work.”
We’re not sure how a model can be “parsimonious,” but, as a taxpayer, would prefer that the Fed be parsimonious with its budget. The New York Fed likely designed the model to mimic “the way markets work,” so it can also be manipulated by the Fed.
Given positive news over the past week about trade data and wholesale inventories, both GDPNow and Nowcast have boosted their estimates of GDP by 0.2% over the past week. Whoopee!
Our conclusions from the latest bit of Fed forward guidance are that:
- The economy remains in rotten shape and is nowhere near the rate of growth it should be at.
- The Fed won’t be raising interest rates again anytime soon.
- New York Fed President Dudley and Atlanta Fed President Dennis Lockhart may not like each other very much.