EDGAReView

We asked our Senior Analyst “EJ”, Elias-John Kies, to do a little showing off for our many subscribers who live inside Excel, and are looking closely at banks. His analysis, updated within seconds of the banks’ disclosures, is just one example of what you can accomplish when your worksheet is directly connected to SEC data, via a pipeline through EDGAR Online. It’s the fastest, most comprehensive, and most instantly re-usable data available. See how fast and deep you can dive into banks’ data.

EJ broke down the financial reports of four companies – Goldman, Morgan Stanley, JP Morgan, and Citigroup using the reports being submitted to the SEC. Here is the spreadsheet he created, using the I-Metrix Professional Excel Add-in and XBRL data from EDGAR Online.  And his explanation is below. Unlike this copy, though, EJ’s own worksheet is live: it is automatically updated when any new data is filed with the SEC, ensuring that his analysis is near real time. [If you want to experience a live spreadsheet, get a free trial of I-Metrix.]

Speed is essential for analysts, but as incredible as it may seem in 2009, most financial data is still created by a labor intensive manual method. Data vendors have groups of people standing by (generally overseas), ready to take the latest SEC reports and manually rekey or copy and paste the line items into a proprietary system where the data is crunched and normalized before it is sold.

The process has been around a while, and while it is fairly efficient for producing Fortune 500 data within 24 hours of filings, it may take weeks or more to be created in any detail for mid-cap and smaller companies, if at all. And those databases are seldom updated when corrections are filed. Most users are not aware of how old their data is, what is missing from the original SEC disclosure, or even know how the data was collected.

In contrast to this manual process, which is prone to error, EDGAR Online automatically extracts entire SEC reports in real time from the SEC EDGAR database and then turns every data point from every-size company into computer-readable interactive data (XBRL). Over 2.5 million business rules are applied to the SEC filings by EDGAR Online in the process of making the data clean and computer-readable. All this is done before passing it on to subscribers.

Along with the most current data, EDGAR Online has a unique, vast database of historical XBRL data from 1999, allowing analysts to research trends for companies and for entire industry segments.

EJ’s COMMENTS ON HIS WORKSHEET:

Shown here as a Google Doc, the analysis has 4 companies (Goldman Sachs, Morgan Stanley, JP Morgan, and Citigroup) with breakdowns of certain revenue components (brokerage, investment banking activities, trading/investments) and expenses (Sales, General & Administrative, and Employee Compensation).

The banks are broken into:

1) Margin % of Revenue for each element

2) Quarter over consecutive Quarter growth

3) Year over year growth (same quarter, previous year)

4) Amount on a per share basis (to display contribution to EPS)

The top of the spreadsheet shows consensus estimate data and how the companies reported Q3 EPS that beat the consensus estimates.

This table details how Goldman specifically blew through the consensus estimates with an Earnings Surprise of  $1.12 per share.

While many data companies can extract EPS quickly, for underlying data elements such as Employee Compensation, extraction of the information could take substantially longer than what is available from EDGAR Online. EDGAR Online extracted the earnings release data (8-Ks)  within 3 hours for most of the companies. This is consistent for all companies in EDGAR Online’s database regardless of size.

An interesting point related to Employee Compensation data is illustrated at the bottom of the worksheet. A major reason why Goldman beat the consensus estimate was due to a reduction in the employee compensation as a percent of revenue from a 3-year average of 48% down to 43%. Factors such as public outrage over employee comp and Goldman’s widely anticipated highest revenue year ever contributed to the reduction in percent of revenue and were particularly useful for PR purposes. It appears that the consensus estimates used the historic average which allowed a contribution of approx. $1.00 earnings surprise.

Naturally, analysts will now adjust their models for this new scenario to reduce the surprise factor’s impact on stock movements. With this timely and granular information, investors may want to reallocate their funds to potentially more lucrative stocks.