How do you calculate terminal value in a DCF?

Terminal value is what the business is worth at the end of the forecast, for all the years I don't project one by one. There are two ways to get it, and I'd run both. The first is perpetuity growth. I take the final year's free cash flow, grow it one year, and divide by WACC minus the growth rate. Say year five free cash flow is $200 million, WACC is 10% and I'll assume 2% growth. That's $204 million divided by 8%, which is $2,550 million. The second is an exit multiple. I take the final year's EBITDA and apply the multiple that comparable companies trade at today. If year five EBITDA is $300 million and the peers trade at 9 times, that's $2,700 million. Either way, that value sits at the end of year five, so I still have to discount it back to today. I'd compare the two results to see whether they're close.

Terminal value is what the business is worth at the end of the forecast, for all the years I don't project one by one. There are two ways to get it, and I'd run both. The first is perpetuity growth. I take the final year's free cash flow, grow it one year, and divide by WACC minus the growth rate. Say year five free cash flow is $200 million, WACC is 10% and I'll assume 2% growth. That's $204 million divided by 8%, which is $2,550 million. The growth rate has to stay at or below long-run growth in the economy. Anything higher means the company eventually outgrows the economy it sells into. The second is an exit multiple. I take the final year's EBITDA and apply the multiple that comparable companies trade at today. If year five EBITDA is $300 million and the peers trade at 9 times, that's $2,700 million. Either way, that value sits at the end of year five, so I still have to discount it back to today. Dividing $2,550 million by 1.10 to the fifth power, which is about 1.61, gives roughly $1,580 million. I'd lead with perpetuity growth, because it keeps the DCF independent of how peers are priced today, and use the multiple as the check. It's just as common to lead with the multiple, and either way the other method is the check. Here $2,550 million over $300 million of EBITDA implies 8.5 times, close to the peers at 9. Going the other way, the 9 times multiple implies growth of about 2.4% a year forever, which I can believe. If it implied 5%, I'd know the multiple was too rich. One caveat. The final year has to be a normal year, with capex close to depreciation and margins I'd expect to last, because any error in it is repeated in every year after.