Bailout

Beware the WSJ?s pay statistics

By Felix Salmon

This is getting to be a habit: today’s WSJ article claiming that Wall Street pay has hit a new record high in 2010 is seriously flawed. “In 2010, total compensation and benefits at publicly traded Wall Street banks and securities firms hit a record of $135 billion”, the paper says — but you have to really hunt to find the small print. First click on the “Interactive Graphics” tab at the top of the story, then click on “About our methodology” at the bottom of that tab. A window pops up in Flash, with uncopyable text saying this:

The analysis includes important caveats. Figures don’t reflect results of acquired companies before they were purchased. For example, BlackRock Inc. bought Barclays Global Investors in 2009, while Bank of America Corp. acquired Merrill Lynch & Co.

It’s not that the WSJ couldn’t have done a proper analysis. Certainly Bear Stearns and Merrill Lynch and Lehman Brothers were an important part of Wall Street compensation in 2006 and 2007, and they made their compensation numbers public. If you want to run the total amount of money paid to employees by public Wall Street firms over time, you have to include the firms which no longer exist.

The WSJ doesn’t do so, however — and, what’s worse, hides the fact that its analysis is woefully incomplete. The problem, I think, is with data providers, who are really bad at providing any kind of historical data for public companies which used to exist but don’t any more. Companies tend to be identified by their ticker symbol, and so when the ticker symbol disappears, so does the data — especially when the ticker is then taken over by someone completely different.

But it reveals little to say that Bank of America or JP Morgan are paying their employees more than they did in 2007, when the real comparison should be with how much Bank of America and Merrill Lynch combined were paying, or JP Morgan and Bear Stearns. (And WaMu, for that matter.) The WSJ has the resources to find SEC filings for old public companies and add them in to its calculations. Not to do so is just laziness, plus a natural gravitation towards the most sensational possible headline.

Full post as published by Macroprudential on February 02, 2011 (boomark / email).