The year 2013 starts with a scandal in the usually buttoned-up circle of auto analysts at America’s brokerage and research houses. For more than a year, TTAC has graded the predictions of analysts, as published by Bloomberg, against (hopefully real) data published by OEMs a few days later. In all that time, we haven’t seen a drama like yesterday: For the second time, Bloomberg ruined the chances of an analyst. Now, the analysts fight back.
On Friday morning, well before January sales numbers were in, an email reached us from Jefferies star analyst Peter Nesvold:
“Bloomberg reversed our Ford and GM forecasts. Our published forecasts were for a 15.3 million SAAR, Ford up 21.2%, GM up 13.7%, and Chrysler up 15.8%.”
“Reversed” was putting it nicely. “Screwed-up” would be more like it. Nesvold attached a research note, published on January 29 that backs up his statement. (Download research note here.) This was not the first time that Bloomberg messed up the numbers. This time, there was still time to do something about it.
|Top Ranked Analysts, January 2013|
|Rank||Analyst||GM||Ford||Chrysler||SAAR||SAAR Diff||OEM Diff||Overall|
|1||Peter Nesvold (Jefferies)||13.7%||21.2%||15.8%||15.30||0.07%||3.31%||3.37%|
|2||Jessica Caldwell (Edmunds.com)||16.0%||21.0%||20.0%||15.30||0.07%||4.38%||4.45%|
|3||Joseph Spak (RBC)||14.0%||21.0%||15.0%||15.40||0.72%||4.01%||4.73%|
|4||Jesse Toprak (TrueCar.com)||14.0%||18.0%||18.0%||15.40||0.72%||7.22%||7.94%|
|5||Emmanuel Rosner (CLSA)||13.0%||17.0%||16.0%||15.20||0.59%||8.01%||8.60%|
|6||Brian Johnson (Barclays)||13.0%||18.0%||18.0%||15.10||1.24%||8.22%||9.46%|
|7||John Sousanis (Ward’s)||12.0%||19.0%||18.0%||15.50||1.37%||8.22%||9.59%|
|8||Chris Ceraso (Credit Suisse)||11.0%||17.0%||13.0%||15.20||0.59%||13.01%||13.60%|
|9||Alec Gutierrez (Kelley Blue Book)||7.2%||16.0%||17.0%||15.10||1.24%||15.02%||16.26%|
|10||Patrick Archambault (Goldman)||7.9%||15.0%||11.0%||15.00||1.90%||20.11%||22.00%|
|11||Rod Lache (Deutsche Bank)||7.5%||10.0%||8.0%||14.80||3.20%||28.51%||31.71%|
|12||George Magliano (IHS)||NA||NA||NA||15.20||0.59%||300.00%||300.59%|
|13||Alan Baum (Baum & Associates)||NA||NA||NA||15.20||0.59%||300.00%||300.59%|
|14||Adam Jonas (Morgan Stanley)||NA||NA||NA||15.40||0.72%||300.00%||300.72%|
|15||Ryan Brinkman (JPMorgan)||NA||NA||NA||15.00||1.90%||300.00%||301.90%|
|16||Itay Michaeli (Citi)||NA||NA||NA||15.00||1.90%||300.00%||301.90%|
|17||Jeff Schuster (LMC Automotive)||NA||NA||NA||15.00||1.90%||300.00%||301.90%|
Because the analyst sent in the correct data, along with the research note, before the sales data were published, we decided to use the correct data, and not the bogus data reported by Bloomberg. Using the correct data, Peter Nesvold starts the year as the top analyst, beating last year’s queen of automotive analysts, Jessica Caldwell, by a comfortable margin.
|Without the intervention …|
|1||Jessica Caldwell (Edmunds.com)||16.0%||21.0%||20.0%||15.30|
|2||Joseph Spak (RBC)||14.0%||21.0%||15.0%||15.40|
|3||Jesse Toprak (TrueCar.com)||14.0%||18.0%||18.0%||15.40|
|4||Emmanuel Rosner (CLSA)||13.0%||17.0%||16.0%||15.20|
|5||Brian Johnson (Barclays)||13.0%||18.0%||18.0%||15.10|
|6||John Sousanis (Ward’s)||12.0%||19.0%||18.0%||15.50|
|7||Chris Ceraso (Credit Suisse)||11.0%||17.0%||13.0%||15.20|
|8||Peter Nesvold (Jefferies)||22.0%||14.0%||16.0%||15.30|
Without the intervention, Bloomberg’s bogus data would have landed Nesvold far behind in place 8.
As we said, this was not the first time. After we crowned Jessica Caldwell “Most Reliable Analyst of 2012,” we received an email from Brian Johnson, top analyst at Barclays. Johnson noted that back in April, Bloomberg had peculated his prediction for Chrysler and said he did not deliver one. According to TTAC’s rules, a missing prediction for the Detroit 3 carries a heavy penalty, which landed Johnson in place 11, instead of place 6. The painful part came many months later: Would Bloomberg have reported the correct number, Johnson would have finished the year ranked third, behind Jessica Caldwell in first and Jesse Toprak in second. Instead, and quite ironically, Peter Nesvold could step on the podium.
Due to the fact that we do not want to rewrite history eight months later, and because we did not want to set a precedent, we had to deny a revised table. We take this opportunity however, to set the record straight.
The record needs to be set straight even more so that both in Brian Johnson’s case and in that of Peter Nesvold, Bloomberg never corrected their erroneous report. The table that kicked Johnson off the podium has been there since April, the table that nearly did cost Nesvold his title likewise remains uncorrected.
Attention All Analysts: Now and in the future, we will correct errors in Bloomberg’s table if the correction is sent to TTAC before the respective sales data are published. Please include the research note. Actually, please put TTAC on your distribution list for research notes anyway. Many brokerage houses do it already. We read their research with great interest, and they tell us that they are reading TTAC. Please send a message to editors, we will reply with my email address.
Beginning with the new year, we introduce one small change at Grade the Analysts: Instead of data published by Automotive News, we will use the table published by Motorintelligence. This table rounds to one decimal place instead to the full percent as Automotive News does. This makes a difference as the aim of the analysts is getting closer and closer to the final results.
What we will not change are the rules that penalize an analyst for not delivering predictions for all the D3. Even if a few TTAC watchdogs scream “statistical abuse”, or “completely bogus.” This is no statistics, this is a contest, and those are the rules. If you want to complain, please direct your considerable fervor to areas that need improvement. Such as Bloomberg’s data.