Welcome to The Signal’s weekly “Macro Minutes.” Each week, we analyze the most important communications from a specific region and provide insight based on our quantitative analysis of central banks.
The Federal Reserve and the Bank of Japan held monetary policy meetings this week, both of which Prattle forecast would result in rate hold announcements.
Due to the Fed’s recent sentiment trend* (more dovish than before the December 2015 hike), we correctly forecast that policymakers would hold rates while hinting at a December hike.
With a raw score of -0.56 (residual** 0.05), today’s statement scored slightly more dovish than our statistical projection (-0.41, residual 0.20)…but more hawkish than market expectations***. The slightly hawkish score (residual 0.09) of Janet Yellen’s prepared remarks was also consistent with the narrative of an upcoming rate hike, most likely December. In addition, today’s headlines focused on divisions within the Fed, a perspective confirmed by a growing divergence in sentiment data.
Bank of Japan
Also precisely matching our forecast was the Bank of Japan, which held rates while making an effort to steepen the yield curve through meek stimulus. Our data suggested a wide variance in BOJ sentiment ahead of today’s announcement, and the bank followed through with a mixed approach. The communications released by the bank today alone carried a dispersion of 3.88, and even the range of residual scores–normally much tighter–was 3.78.
Despite the BOJ’s unpredictability, the dispersion of sentiment scores leading up to the meeting pointed precisely toward this muddled policy decision.
* Prattle’s models are based on the historical relationship between central bank language and market reaction, which is used as basis of evaluation for future communications. The scores are normalized around zero and range between -2 and 2, negative numbers indicating dovishness and positive numbers indicating hawkishness.
** Residual scores represent the tone of a communication compared to the rolling, 12-month average for that individual communication type or speaker. Raw scores represent the tone of a communication compared to the average of all communications.
*** As suggested by the mock statement written by Larry Meyer and Dave Stockton, which scored -0.71 (residual -0.10)