DCF and NOLs

yig's picture
Rank: Senior Orangutan | 485

I'm creating a DCF for a company that I've been looking at. It has a large nol, and won't be able to use the entire figure in the projection period, unless I stretch it out to a period where my projections will be completely off.

One suggestion I've been thinking about is to do the DCF normally, and then find the PV of the nol.

Does anyone have any other suggestions, and what discount rate would you use to discount the PV tax savings on the nol?


Comments (17)

Sep 28, 2015

I encountered a very similar issue the other day, and decided to take the PV of the nol and add to the DCF value as you suggested. Curious what others think/have done in the past.

Sep 28, 2015

I would not use a DCF. This might not show the true value of the firm. The whole value of the firm would be coming from the terminal value which is too risky and weird.

Sep 28, 2015

How would the whole value of the firm come from the terminal value? I'm still accounting for taxes if I do a normal DCF, and I still have positive FCFF.

What I'm doing is valuing the nol separately, and then accounting it in the same way as cash.

How would you do it?

Sep 28, 2015

Run the nol for the DCF. Build a quick NPV calc of the nol balance past the discreet period. Pay attention to section 382 limitations if applicable. Macabacus has a good template for 382 NOL's.

Sep 28, 2015

From what I understand, this is what you're trying to say.

Do a DCF and apply a 0% tax rate to EBIT.

Since the nol would still be used after the projection period, the terminal year FCFF would essentially not account for any tax paid. How could I make sure that the terminal value accounts for the fact that the company will pay for tax sometime after the projection period ends?

Sep 28, 2015

You can model out actual nol usage using the EBIT, have it set up to apply the tax rate to the taxable income in each period. The terminal period should always include taxes. Any remaining nol after the discrete period should be separately valued and added to the DCF. Does that make sense?

You can also run the DCF with taxes and use something like this model to value the nol separately.

It's really just how you want to present it. Personally I prefer to model out the NOL in the DCF and tack on any remaining after the discreet period.

Sep 28, 2015

It will probably make the most sense to do the DCF without the nol, and then add on the value of the nol on top of your DCF value. Partially because in an acquisition scenario, the NOL value will go down due to section 382 limitations stretching out the realization horizon (ignoring changes in WACC), and so it's important to understand the value of the core business.

You'll use different discount rates for the two: WACC for DCF and Cost of Debt for NOL (search around for explanations why).

I'm guessing you're doing this from a public investment perspective? If so, section 382 limitations would only apply to existing NOLs to the extent the company is limited by how much it can currently use. If this is from an acquisition scenario, the DCF should still be computing the value of the standalone entity. If not, you'll need to use a different WACC (or if it's an LBO, you could flow through NOLs to your CF statement and calculate section 382 limitations based on the purchase price).

Sep 28, 2015

Do a normal DCF. Use a normal tax rate (e.g. 35%). Calculate enterprise value in the normal way.
Then, on top of your standalone calculated enterprise value, add the PV of the nol.

The PV of the nol should be calculated by going out as far as necessary to capture all of its benefits (might need to PV 20 years of tax savings even if the DCF is for 5 years)

Reason I suggest this is that trying to model tax savings into your DCF can be tricky. Requires things like normalizing FCF in your terminal year (which others have correctly alluded to). Also if you are using a discount rate other than your WACC for your NOL, it makes sense to value it separately instead of integrating it into your DCF, since you would effectively be discounting tax savings at different discount rates.

As for the correct discount rate to use for an NOL balance, opinions vary greatly. I've spoken to many intelligent people on the matter and have heard: Ke, Kd and WACC, all with intelligent reasons behind them. But I think Macabacus puts it best: "The discount rate used to calculate the PV of NOL should reflect the probability that a company can generate sufficient taxable income in the future to utilize the NOL; otherwise, the NOL will expire unused and worthless". If a company is nearly certain to use all of its NOLs, ~Kd makes sense. If it is highly speculative, ~Ke makes sense. Anything else would fall in between. WACC is normally a good place to start.

"Success means having the courage, the determination, and the will to become the person you believe you were meant to be"

    • 1
Sep 28, 2015

Great explanation.

Just a few questions. Wouldn't you subtract the PV of the nol from the calculated enterprise value? The way I understood it is that you should treat the nol as essentially cash.

Also, when looking at comps, would you subtract the value of the NOL from EV? The other comps don't have the NOL, and I feel that it's important to account for the NOL somehow.

Sep 28, 2015

No, you are valuing the cash flows of the business. In regards to the nol, the positive cash flow the business receives due to the reduction in taxes. It should be added to the overall PV of the DCF to arrive at EV.

Sep 28, 2015

Mephistopheles, while pre-nol comps say 8x, maybe post-nol comps say 6x, and that's what you should really be looking at to compare companies that have different NOL balances (it's no different than how you treat cash). Whether you call it Adjusted EV or not, your multiple should be post-NOL. If you don't feel that public equity markets are accurately valuing the NOLs, then you should buy the stock because fixing EV will imply stock appreciation to account for the realization of NOLs as additional cash source.

Sep 29, 2015

Yes, if the pubco has significant nol's you can take that into account for the multiples, but you should do that to the comparable company, not the subject. Trying to do that on transaction comps is a useless exercise as the data is going to be unavailable 90% of the time.

Not to mention it sounds like a distressed situation and earnings multiples mean crap in DM&A.

Oct 1, 2015