DCF Vs. Relative Valuation

The short answer to this question is that there is no "correct" answer, and I think that the question's framing is misleading. At a high level, the purpose of relative value is to serve as an approximation of value. The purpose of a DCF is also to serve as an approximation of value. Relative value is often thought of as a shorthand way to get at a DCF value - multiples are typically driven by a combination of growth, free cash flow conversion, leverage profile, and business sustainability. Those are the same factors that drive a DCF. In terms of the idea of having no comparable companies, you'll see below that you pretty much need to take a view on companies to anchor off of to do much of any useful analysis.

In terms of how you want the answer to ultimately shake out, it really depends what you are trying to do this exercise for. Is this for a sellside bake-off? A buyside advisory assignment? Actually evaluating whether to purchase a company as part of a private equity sponsor or a corporate acquirer? And what do you really mean when you say "predictable"? Is it a super early stage company? Or a mature but volatile company?

As a general framework, here is how I would think about it. You will notice that this is not a clean answer, because there is no clean answer. These are the situations where creativity can actually come to bear, and putting real thought in can actually be value additive to a client (as opposed to just running through the motions on straightforward analysis to reconfirm an answer the client has already figured out). 

The first step would be to get a high level sense of what projections could look like. That might require you to think about a few different states of the world, which may in turn depend on the type of company you are looking at. For example, if it is a commodity-driven company, you'll need to figure out the right commodity prices to focus on and then figure out how the projections might look in a low price case, base case, and high price case. If it is an early-stage, consumer-focused company, maybe you need to think through what expenses look like over the next few years and then figure out what an appropriate adoption curve could suggest. To get to an appropriate adoption curve, perhaps you need to get creative around what somewhat similar companies have experienced (for example even if they sell a different product, maybe it makes sense to see how long YoY growth doubled for iPhones to get an idea of how long it could last for Peloton). One way or another, you need to at least have some idea of what projections could look like. At some point, that is going to probably require analogizing to somewhat similar companies (again, from a growth driver perspective, even if the specifics of the businesses are very different).

The above exercise will naturally end up requiring you to take a perspective on a very loose comp set, which will now give you tools to run relative analysis, as well as DCF analysis. The original post misses the fact that to run a DCF, you will need some idea of comparable companies, otherwise you have no way to figure out the appropriate cost of capital for the business you are evaluating. Again, even if you land on very roughly analogous comps that operate in different sectors at a granular level, the goal is to find companies that are subject to similar growth drivers, demand dynamics, etc... You may very well end up with three, four or five buckets of comps that each match up well with the business through certain lenses, but less so through others (for example, one comp set may be similar from a gross margin profile but slower growth, another comp set may be similar from growth but have much lower margins).

From there, you can start thinking about what the different analytical approaches give you. Do most of your comps trade on AV / EBITDA? AV / Revenue? P/E? This is where a football field can actually start being helpful. Run the right metrics on the right comp sets, spread across your base, low and high case. Run a DCF based on costs of capital that align with the full comp set, and then perhaps a narrower comp set that you think has the best applicability to your business. Think about who the right buyer for this business is. If it is a sponsor, what type of sponsor? One that typically demands 25% returns? Or 10% returns? How does that line up with the cost of equity that you have arrived at based on your comps - does it feel defensible? If it is a strategic, what strategics? What is their cost of capital? How do they typically evaluate targets (if they have publicly disclosed it)? Maybe it makes sense to run an ability to pay analysis for the three strategics that would be the best buyers here (whether that is looking at accretion / dilution on the most relevant metric, or applying whatever other framework they have publicly pointed to as their evaluation method).

Once you have done all of this, you can start to sanity check. Do some answers differ wildly from others? Why is that? Maybe the cost of capital for one comp set is way below that of another, even though they seem to have very similar business profiles. Is that because one has an ESG halo and so public investors have bid up their values in a way that the other comp set hasn't seen? Which better lines up with your company? And again, what are you trying to tell the client? Are you pitching a seller on how you would try to frame an aggressive sales case to get buyers to use the most favorable peers? Or are you helping a buyer think through the most reasonable price to pay? That's going to result in different answers.

Unfortunately, there is no one-size-fits-all answer to this question, because it really is the type of situation that settles in the cracks of standard banking assignments. Hopefully it's also the type of thing that you actually find somewhat interesting, because it is among the more "fun" things to actually try to think through (at least when you have the time, though that is obviously at a premium for most people right now unfortunately). So while the above won't give you a definitive answer, hopefully it provides some food for thought and potential ways to approach the question.

It's also worth noting that while an analyst can and should think about these and try to come to a reasoned perspective on some of these questions, I would never expect an analyst to get this 100% right (or even 50% right) - I would expect them to understand the goal, help pull together the right information to help us think through this, and take a crack at some perspectives on solutions. Then, I would expect them to be curious on how I think about it, ask questions about my reasoning as I walk through what I think we should do, and not get defensive if I ultimately don't agree with the approach that they proposed. The key is the curiosity and the solutions-oriented approach.