More than a month ago, we reported on the record steepness of the yield curve. The implication, according to the Fed’s own research on the forecasting properties of the yield curve, is that economic growth might turn out to be pretty strong.
Today was the release date for the first cut at Q4 GDP growth. GDP growth in Q3 was already positive at +2.2% and the consensus estimate of economists for Q4 was a significant acceleration to about +4.7%.
Maybe the yield curve knew something the economists did not because the first report for GDP came in at +5.7%, much stronger even than the already significant acceleration that was predicted. (See here and here for two perspectives on the GDP report.)
Now, a lot of brainpower goes into GDP estimates. Literally every macro economist is fiddling with a spreadsheet trying to figure out what the proper inputs are. The consensus number, which over time tends to be more accurate than the forecasts of any individual economist, is simply the average guess of lots of highly educated guesses. Even so, it turned out this quarter that their guess was far short. Economists who were giving significant weight to the shape of the yield curve and other market data, on the other hand, had much higher estimates of economic growth.
It gets back to the issue of making decisions based on expert judgment or based on piles of market data. In most arenas, while expert judgment is certainly superior to amateur’s guesses, generally basing decisions on market-generated data is the way to go. Data is unemotional and has less bias than judgment calls. It’s also the reason that our systematic relative strength process is run exclusively on market-generated data-price.
Market reaction to the GDP data will be interesting. Will the market be positively surprised by the GDP data and melt upward as Bill Miller predicts? Or will we discover that the market already had strong GDP estimates baked into its current price and nothing much will happen? We’ll just have to wait and see, but the primacy of using market data for decisions over more judgmental methods of forecasting has been reinforced once again.
“It gets back to the issue of making decisions based on expert judgment or based on piles of market data.”
Yep. And the “expert judgment” is that everybody in bonds should be short duration into rising rates.
Meanwhile, have you looked at the market-based data concerning what typically happens over the 1-3 years following a rate bottom with an extremely steep curve?