Expected Values and NPV for Capital Projects

Replies to This Discussion

Permalink Reply by Emre Ozmen on May 5, 2010 at 9:17am

I think this issue is rather open to different opinions. But in expected value cases, its usually the probability * return= EV.

But, probability*return or expected return is multiplied together because that specific probability is the probability of the amount of that return.

For example: Lets say you have an investment with three return assumptions: 10M, 20M, 30M with probabilities 50% 30% 20%:
the expected value of that would be:

10*.5+20*.3+30*.2= 5+6+6=17M would be the EV of that stock.

However, in this case, the probability you are taking into account refers to the probability of you winning thebids for the 3 projects. So in this case i would say that you wouldnt account for the probability of winning the bids in the cash flows, but you would use a specified discount rate say 9% for those cash flows.

What you could use the probability of winning the bids for would be to assess the amount you would want to bid for each project.

I too am a little rusty on this case, but I would go about it like this.

Id use the probabilities of winning to determine my bid amount, because i would assume that the company has a certain cap to invest.

• ▶ Reply

Permalink Reply by Russ.F on May 5, 2010 at 2:38pm

To answer your final question, you should use probability distribution on the NPV/IRR and not the actual cash flows related to the project. I think this is also validated by Emre's response in that you should be applying probability to the expected return (NPV/IRR).

Hope this helps.

• ▶ Reply

Permalink Reply by Emre Ozmen on May 6, 2010 at 9:48am

Yup, that makes completely perfect logical sense. I would also apply it to NPV IRR and not cash flows directly

• ▶ Reply

Permalink Reply by Gregory Odum on May 10, 2010 at 9:35am

The best method would be to use Real Options in valuing the project, to take into account the volatility in the expected value.

• ▶ Reply

Permalink Reply by Amara J Moosa on May 18, 2010 at 9:48pm

I agree. In the real option scenario, he can take into the consideration of the probabilities of not receiving the actually cash flow. Attached is a sample from one of my old corporate finance classes.

Attachments:

• Chapter 13 Solution.xls, 27 KB

• ▶ Reply

Permalink Reply by Umed Saidov on May 10, 2010 at 9:35am

I agree with Emre in that the probability of winning projects is separate from the probability projects' CFs. These are two separate probabilities and should be treated as such. What I am not sure about is the conclusion (if I understood everything correctly):

My take is that one can't quite use the probability-weighted investment amount to bid for the project as it may affect the outcome. The actual probability of winning projects (apart from number of participants) includes ranking of bids by amounts. In that respect, including probability of winning into your bids may hurt your chances of actually winning the project if your bid comes higher (all else equal).

Instead, it would make more sense to use the actual investment amounts. This way you keep the risks/rewards of the project separate from the probability of winning them (all else equal).

• ▶ Reply

Permalink Reply by Christopher Smith on May 10, 2010 at 11:12am

Discounting the revenue is not the right way to go about this. Your projected revenue would not be reduced based on the number of people bidding on the project.

I having two bidders does not give your a 50% chance at winning. Your cost & services offered/provided to your customer are going to be the things that drive your probability to win the bid.

If you are way out of line with your competitor you have a 0% probability of winning the bid. (i.e. if you give your business away you have a 100% chance of winning the bid if you are 50% more than market/competition you typically don't have a shot at winning).

I would model your projects on different proposals from what you think would be a slam dunk at winning the bid to what you know would be way out of the game. By plotting the NPV/IRR vs your assumed probability will give your a good idea of the better project. IT will also tell you at what price is the lowest you can go to have acceptable returns.

• ▶ Reply

Permalink Reply by Russ.F on May 10, 2010 at 12:46pm

I agree to each point made above and something Umed pointed out that I didn't pick up on was the use of this technique in "choosing" the project. My response assumed multiple outstanding bids and the ability to project the expected return related to such outstanding bid's.

The outstanding bid's would require a probability distribution of expected outcomes to risk adjust your forecast (all other things equal). And I will second the notion that probability is not correllated to your bid's percentage of total bids (on a bid-by-bid basis). You should utilize average outcomes from historical bid's adjusted for pertinent variables (i.e. market power, competitor power, long list of etc's.)

• ▶ Reply

Permalink Reply by Frederick H on May 10, 2010 at 4:59pm

The NPV of each projects is the NPV of each project. The expected value of each of your options is the expected probability times the NPV of each project.

• ▶ Reply

Permalink Reply by haresh shah on May 11, 2010 at 6:44am

I think question is not raised to get the right direction in the matter. The ral question should be whether i have enough alternative oppotrtunities or not ? if not,i must bid for both the projects. If There is surplus capacity and the company continue to incure fixed costs ,it is better to work toward s improving probability of getting aproject instead of going into finacial calu lations

• ▶ Reply

Permalink Reply by Derek Toth on May 11, 2010 at 9:51am

You should adjust the cash flow by the probability of it occuring to calculate you NPV/IRR ($100 mill with 50% chance = $50 mill expected cash flow for NPV calc). This can be done using risk adjusted NPV calculations or decision trees. This type of thing is done in the pharma/biopharma as well as oil industry all the time when there is uncertainty about future cash flow (drug doesn't make it to market or oil well is drilled but no oil is found).

• ▶ Reply

Permalink Reply by aljatar on May 11, 2010 at 1:17pm

Hi

I believe that if you stick to DCF valuation, i believe that you cannot embed the risk in your cash flow forecast but only in your discount factor?

I think that you should identifya discount factor per project and weight the aforementionned discount factor by the size of the associated project to find a kind of WACC.


I am not sure that Real options are relevant here since the use of RO embed the flexibility and I feel that you do not own the flexibility but you rather live the flexibility of others (the bidders)


Hope this was clear,


Please apologize for my poor english, I am not sure my explanations are clear.

Regards,

• ▶ Reply

• ‹ Previous
• 1
• 2
• Next ›
• Page