Welcome, Guest

Calculate Return on Capital
(1 viewing) (1) Guest

TOPIC: Calculate Return on Capital

Calculate Return on Capital 1 week, 1 day ago #2667

  • Pottz
  • OFFLINE
  • Fresh Boarder
  • Posts: 14
  • Karma: 0
I am in the process of reviewing a data set extracted from Bloomberg with that from Screener.co in order to transition to using screener.co as my data source.

The data point I am reviewing currently is Return on Capital. The field source in Bloomberg is RETURN_ON_CAP (as opposed to RETURN_ON_INV_CAPITAL (ROIC)), so I am presuming it is ROCE I am looking to match.

I have tried the function ROCE = EBIT / Capital Employed, where Capital Employed = Total Assets - Curr Liabilities
by creating an advanced custom variable with the function:
aebit/(acurast-acurliab)
(Note I strip out the curr parts to keep it simple and should not impact the ratio)

I do not get numbers that look even close, and I get no results for banking stocks.

An example of a non banking stock would be:
ASX:WES ROCE calc as above in Screener.co = 3.7783, Bloomberg reported number is 14.72
(last FY= 12.13, and ROI from Screener.co = 10.37 (I was looking to see if ROI got closer numbers)

Do you have ROCE or ROC in the system? IS the above approach sensible for use with the screener.co data? Any help would be much appreciated.

Thanks

Steve
Steve

Re: Calculate Return on Capital 1 week, 1 day ago #2670

  • arturop
  • OFFLINE
  • Fresh Boarder
  • Posts: 4
  • Karma: 0
I think the problem is with the "acurast" bit, as that would represent Current Assets. If you use "ata" for total assets, last fiscal year you get 16.52 which is not the same as Bloomberg but much closer

Re: Calculate Return on Capital 1 week, 1 day ago #2671

  • Pottz
  • OFFLINE
  • Fresh Boarder
  • Posts: 14
  • Karma: 0
Thanks for your input, I'll make that change and see how it looks with the other stocks.

Is there an easy reference to the underlying field names somewhere? I have to keep adding conditions using the Basic Screener, then looking at the detail under the advanced Screener to derive the underlying field names - quite time consuming.

thanks again
Steve

Re: Calculate Return on Capital 1 week, 1 day ago #2672

  • Pottz
  • OFFLINE
  • Fresh Boarder
  • Posts: 14
  • Karma: 0
Hmmm, the above suggestion has not helped much. This is a sample of the data I am comparing:

Stock Bloomberg Screener.co
RETURN_ON_CAP aebit*100/(ata-acurliab)
BHP AU Equity 13.62 18.03
RIO AU Equity 13.42 21.84
CSL AU Equity 19.85 21.81
FMG AU Equity 29.73 33.44
WOW AU Equity 8.52 11.27
WPL AU Equity 1.53 6.37
WES AU Equity 14.72 16.52
QAN AU Equity 0.63 -0.73
NCM AU Equity 7.56 9.42
BXB AU Equity 9.76 14.23
S32 AU Equity 1.57 1.93
TCL AU Equity 1.66 2.03
AGL AU Equity 5.78 10.77
SYD AU Equity 3.34 8.13
STO AU Equity 5.47 8.57
ORG AU Equity 0.53 3.26

Anyone have any ideas where I am going wrong here??
Steve

Re: Calculate Return on Capital 1 week, 1 day ago #2673

  • arturop
  • OFFLINE
  • Fresh Boarder
  • Posts: 4
  • Karma: 0
I have tried a few, the original one (the last one below) seems to be the closest to your figures:

Net Income available to common-most recent fiscal year / ( Assets, total-most recent fiscal year - Liabilities, current-most recent fiscal year )

Earnings after taxes-most recent fiscal year / ( Assets, total-most recent fiscal year - Liabilities, total-most recent fiscal year )

Earnings After Taxes, Normalized-most recent fiscal year / ( Assets, total-most recent fiscal year - Liabilities, total-most recent fiscal year )

EBIT(A) / ( Total Assets(A) - Total Current Liabilities(A) )

For me, the first one above matches quite close with another screener ROCE, this one based in Income After Taxes / Common Equity. But doesn't match your Bloomberg's figures.

Re: Calculate Return on Capital 1 week ago #2674

  • admin
  • OFFLINE
  • Administrator
  • Posts: 254
  • Karma: 12
I'm not sure. I tried Googling RETURN_ON_CAP from Bloomberg to try to find a data definition, and was unable to. It is unclear exactly how they compute their metric.

Do you know if they are using fiscal year values or trailing twelve months? All the formulas listed above use most recent fiscal year metrics. It is possible that they are using TTM.

Another potential issue with the formula:

aebit*100/(ata-acurliab)

You probably want to surround aebit*100 with parenthesis:

(aebit*100)/(ata-acurliab)
Time to create page: 0.17 seconds