NPV calculation seems to disagree with IRR calculation on break-even point

Hi Paul,

 I have another question for you, this time around IRR & NPV.

I am currently trying to determine what the PPA price would be for a given system, assuming that the developer just broke even (NPV = $0). Then, I'd like to be able to model a developer real discount rate (at, say 6.1%) and see what PPA price would be required to hit that return.

This seems like a perfect fit for the 'Specify IRR target' mode. If I set the IRR target to 6.1% and the real discount rate to 0%, I should find the first year PPA price that exactly gives an NPV of $0. (This works, and gives me a PPA price that looks reasonable) I can then check my work by:

1) Switching to 'Specify PPA price' mode.

2) Setting the PPA value to the PPA price I just found.

3) Setting the real discount rate to 6.1

This works as well, and I see an IRR of 6.1% and an NPV of $0, indicating that it just equals the hurdle rate for the developer.

However, in the real world, we have inflation and a PPA escalator. If I set these to any value, either separately or the same value, the process above does not hold. The PPA price returned by IRR mode does not result in an NPV of $0 when I use that PPA price (I tried first-year, real, and nominal values) to check using the discount rate.

For instance, in the attached SAM file, I had set both Inflation and the PPA price escalator to 1%, and the real discount rate to 0%. When I use the 'Specify IRR target' mode, I get a first year PPA price of 6.01. If I then set switch to 'Specify PPA price' mode, set the PPA price to 6.01 and the real discount rate to 6.1%, I would expect to get an NPV of $0, and an IRR of 6.1%. The results: IRR = 6.1%, but the NPV is about negative $4 million!

I'm fairly confused here - I thought I understood what SAM was doing, but I am not sure now. I've read the 1995 paper on the various financial metrics but am still flummoxed. What's going on here? Additionally, what is the best way to reliably find out the break-even price (NPV = $0) for a developer, both with a 0% IRR return as well as an arbitrary return (6.1% in our case)? That is the real question I am intending to answer.

Thank you,
Colin Schimmelfing

Hi Colin,

SAM's "Internal Rate of Return" Help topic (look under "Financial Metrics" in the table of contents) defines the IRR as the nominal discount rate that corresponds to a net present value of zero.

For the Single Owner financial model, if you set the target IRR on the Revenue page to the nominal discount rate from the Financial Parameters page, you should get an NPV of zero, as long as you also set the IRR target year to the analysis period.

Best regards,
Paul.