I would like to get input on how to correctly implement Return on Invested Capital in P123. I have found two general formulas for it and have converted to what I think are the right factors in P123. Which do you think is more accurate. Any comments are welcome.
General equation:
ROIC = ((1 - tax rate)(EBIT)) / ((Total Assets - cash) - (non-interest-bearing current liabilities))
In P123:
(OpInc*(1-(TxRate%/100))) / (AstTot-CashEquiv-Payables-TxPayable)
ROIC = ((1 - tax rate)(EBIT)) / (Total Long Term Debt + Shareholder Equity)
In P123:
(OpInc*(1-(TxRate%/100)) / (DbtLT+EqTot))
I saw OpInc is recommended over EBIT.
NOPAT is OpInc*(1-(TxRate%/100))
CashEquiv includes short term investments but I could not find in P123 what handles the long term investments so it is not included.
Non-interest-bearing current liabilities (NIBCL) are for:
accounts payable that have no associated fees or interest
taxes that have not yet been paid and are not increasing because of penalties or interest
current income taxes that must be paid by the end of the year
I used payables and Txpayable for NIBCL.
I think I mapped it correctly but…
I like the first equation better. Both simulate about the same.
Have to be careful with big formula in ranks. They can do things we don’t expect.
One thing I don’t like about this is that it’s rewarding a company for having both cash and debt.
I might want to reward cash, but I don’t want to reward debt. But because it’s a ratio and you are subtracting debt from the denomitor…
More debt means a smaller denom, which means a bigger ratio. Two identical companies with same OpInc, same tax rate and same assets and you will choose the one with more debt. That doesn’t make sense to me.
So…I might try something like this:
(OpIncTTM*(1-(TxRate%TTM/100))-DivPaidTTM)/ (AstTotTTM-CashTTM-CashEquivTTM)
Then create a different ratio that looks at debt. And make that a separate factor. Complex rank formulas can have unintended consequences.
Maybe add this formula:
(OpIncTTM*(1-(TxRate%TTM/100)))/ (DbtTot
TTM+TotEquity)
Then you equal weight both, or play with weights, in a node. Can do the same with CurLiab in denominator. So looking at OpInc/Assets and Opinc/Debt and finding company that’s doing well at multiple ratios - and make this a composite ranking node.
Tom,
Your feedback about the effects of complex formulas in ranking is appreciated. I will experiment with splitting up per your suggestion. The origin of the formulas came from others. As I understand, the notion of including debt was to insure one saw the return based upon the total assets at hand and not pick and choose. this is supposed to give a more complete idea of management effectiveness than what is used in ROE, for instance. I will continue looking at this with your suggestions.
I think your 2nd formula approximates ROIC better. But it is just an approximate value for ROIC.
ROIC = ((1 - tax rate)(EBIT)) / (Total Long Term Debt + Shareholder Equity)
In P123:
(OpInc*(1-(TxRate%/100)) / (DbtLT+EqTot))
Invested Capital should exclude Excess Cash but the entire cash. Excess cash can be estimated (according to McKinsey: Valuation, 5th ed.) as anything over 2% of sales.
For Invested Capital, we are trying to measure the investor’s capital to fund operations. Investors being both the equity and debt holders. It is worth noting that ROIC is the same as RNOA (Return on Net Operating Asset). So if you can figure out how to code NOA-NOL, you can also calculate ROIC.
I am very new to P123, so I am still learning how to code.