I was not able to find a description of this. I assume this is a projection of EPS growth. Can anyone clarify this? Thanks.
Bill
I seem to remember it being a 3 to 5 year estimated EPS growth rate. The data comes from Compustat, so it is unclear whether it is the GAAP or non-GAAP version of earnings growth being estimated. This is a problem when analysts are projecting cash flow or operating income versus earnings. It is also a problem when the EPS baseline is negative. If you want, I cod try rooting around in Compustat’s data dictionaries for a better answer.
Thanks for the info. If it wouldn’t take too much it would be helpful to understand this a little better.
You aren’t going to get help there. It’s not a Compustat number. Estimates is a separate database and Compustat takes it in from another source, and even they won’t know. It’s really something filled in by the analysts; or at least those that choose to publish such a projection.
It’s a soft number that means different things to different analysts and investors. I refer to it as 3- to 5-year out of habit from when I was at Value Line, and from what I’ve seen elsewhere, many others have this framework in mind (probably because at one time, many on the Sell Side came from Value Line or were hired into the Street by ex Value Liners). It’s GAAP EPS (although you should consider it an operating number – analysts don’t generally project unusual gains/losses).
As to a negative base, again, this is a soft number. The best you can do is presume the growth rate starts from a point that represents a return to normal profitability.
This is not a datapoint you can use if you want to go strict quant. It can be a good number to show sentiment (revisions), depth of coverage (prevalence of LT forecasts relative to current forecasts), or PEG comparisons (whatever deficiencies LT growth has for use in PEG, another soft number, it’s a heck of a lot better than any historical growth rate one might choose because at least LT represents someone’s view of something more normal. And because it tends to be more normal than precise, it’s much better than a growth rate for use if you want to create a formula that needs to step up EPS into the future (but you’ll need some screening/buy rules to weed out unusable base points).
When using LT growth rate, do so with the following Fisher Black quip in mind, something he said after moving from MIT to Goldman Sachs: “Markets look a lot more efficient from the banks of the Charles than from the banks of the Hudson.” My alternative: Reasonable approximation trumps false precision.
Thanks for the thoughts Mr. Gerstein.
Seems like a problem that there is so little transparency on estimate data. I rely heavily on estimates in my quant models. I unfortunately have to accept their imprecise nature in order to capture a forward looking valuation.
Marc may be correct about the “fuzzy” nature of estimate data. However, I was hoping for something more concrete and feel that more certainty can be had with a little bit of digging. I am going to rely on the Compustat Data Guide, available at: http://finance.fbv.kit.edu/rd_download/NorthAmerica_DataGuide.pdf. Even if there are differences between the data shown in the guide and P123’s data subscription, it shows that S&P’s vague answer regarding the source and nature of estimate data is unsatisfactory.
In reference to the definition of the analyst’s long term growth rate, it appears that a 5 year timeframe is appropriate. There are two data elements which seem relevant: ‘FCSGRAEC’ and ‘IBLGAC’. ‘FCSGRAEC’ is the “Thomson First Call Mean - Growth Rate” which is described as follows: “This item represents the latest available mean annual growth rate of all individual analyst growth rates.” ‘IBLGAC’ is the ‘Thomson I/B/E/S 5 Year Growth Avg. Est. - Current’ which is described as follows: “This item represents the latest mean or arithmetic average estimate of long-term (5-year) growth for the current fiscal year. It is expressed as an annual growth rate.”
Clarity on whether the P123 estimates are from First Call or I/B/E/S or are amalgamated would be really important to know due to key differences in the datasets. I hope that additional clarity into the origination of data might also lead to more granularity in the P123 interface. For example, the data guide indicates that FIRST CALL differentiates EPS and Cash Flow estimates. Conflating the two would bias the estimate data more than necessary. Additionally, median long term growth estimates seem to be available in both the FIRST CALL and I/B/E/S Estimates. For example, FIRST CALL has a data element called “FCSGRMD: FIRST CALL Median Long-Term Growth Rate” and I/B/E/S has “IBLGMC: Thomson I/B/E/S 5 Year Growth - Median Est. - Current”. The guide suggest that more insight into the data should be possible.
Given this information, is there any way to re-approach S&P Capital IQ to ask for more clarity?
- David A
Marc, thanks for the clarification (GAAP EPS) and insights. I feel a little more comfortable in how to use it.
David, thanks for investigating and pushing this forward; given the potential differences, additional clarity would be valuable.