What is the Right Multiple to Pay for a Stock? (Free Excel)

Get smarter on investing, business, and personal finance in 5 minutes.

This week's newsletter is an adaptation from my recent YouTube video on What is the Right Multiple to Pay for a Stock.

One of the most important questions you can ask as an investor is what is the right price to pay for a company, which is the same question as asking, what is the right earnings multiple?

Because valuing a company on its multiple of earnings is the easiest and most common way to do any sort of valuation.

But there's a question that always lingers and is essentially never answered: what is the right multiple to pay for a business?

I am going to show you how you can go from a discounted cash flow model to a multiple, and vice versa, and the assumptions that are embedded in various multiples, which is absolutely key.

A lot of times investors are unaware of what needs to happen in the future for the investment they're currently making to make them money.

Why are you paying a 20x multiple for a stock?

20x isn't usually thought to be a very high multiple of earnings.

But if you buy a company for 20x earnings, do you know how long it will take you to just get your cash back, if earnings are growing 7%?

13 years.

It will take you 13 years just to get your money back, and then you only start to make a return on the cash flows thereafter.

So why are investors paying these high multiples?

I'm not even talking about Palantir's earnings multiple of 147x, which we talked about in the last week. I'm talking about a stock market average multiple of 20x.

Why do you pay that at all?

In this week's Five Minute Money, I'm going to show you how an earnings multiple maps to a DCF.

Making Your Assumptions Explicit.

In last week’s Five Minute Money, the big point I wanted you to take away was that anytime you pay a multiple, there are implied assumptions in it that need to happen in order for you to make money.

Using Palantir as an example at $172, revenues needed to grow from $6bn to $37bn just to rationalize the current stock price, and to make money beyond that, it has to grow further and still be growing fast.

A lot of times people talking about growth investing and high multiple stocks are coming from this position that you can't actually value a company that is growing very quickly, because the math shows it's so much earnings that it's unrealistic, and on a deep level you don't want to admit, you're a little bit of a momentum trader.

You want to say you're a long-term investor, but you don't fundamentally know what needs to happen for the business to make a return, and you want to stay unaware of that, because once you become aware, these assumptions are going to seem so lofty that you won't end up owning the stock.

I believe that is the case for many people, because there are stocks that do not seem to trade on fundamentals for a long period of time, and then sometimes those same stocks suddenly do trade on fundamentals because they grow so much more rapidly than people expect.

I think Palantir is an example of this.

Even when it was $10, $20, $30 a share, many people said that's overpriced.

But if you paid $30 a share with a $72bn market cap, and right now we're looking at a market cap of $400bn, plug in $30 divided by $166, times $400 billion, that's a $72 billion market cap with $237mn in net income. Whereas today, in the last 12 months, they have net income of about $3bn.

Paying $72 billion a couple years ago, thought expensive at the time, now looks cheap in retrospect, because at the time earnings were 90% less, so that 35x multiple was really several hundred times.

The pushback is commonly, “These assumptions are too lofty, but I'll sell this stock to someone else at 40x earnings anyway.”

My pushback is that when you sell at 40x earnings, there are already assumptions of growth embedded in that.

Whether you're aware of them or not doesn't matter, that's already factored into the multiple.

If we look at SpaceX stock, they are trading at a $2 trillion valuation.

And a lot of people want to own it just because it's Elon Musk, without doing the precise math of how much revenue is needed to rationalize that valuation.

Elon Musk actually does talk about this on the earnings call.

He says they're projected to reach a trillion dollars in revenue, not in 2031, now in 2030.

That means in 2030, it's going to be trading at 2x sales.

For the last 12 months, revenue is $23bn, and he thinks they'll hit $100bn this year, so even after growing 4X from LTM, they'll grow another 10X.

This is huge growth, but it's the growth actually required to support a valuation like this.

There's going to be people that invest in a stock like SpaceX not because they believe a trillion dollars in revenue, they never did that math, they just believe in Elon Musk.

You get this criticism for Tesla too.

Some investors invest in public companies more like a VC firm would, without a great sense of the numbers, just trusting the managers.

Because it's not tied to valuations or projections, there's a much higher possibility the valuation becomes unmoored from what's practical.

That doesn't mean valuations will never matter, they just won't matter so long as most people who own the stock don't care, because a stock's price is the perception of value, and perception is always subjective.

It's not how I would invest, because I want earnings downside protection, actual cash flows from the business.

But different ways of investing work for different people.

This is kind of what happened with Adam Neumann in WeWork.

He spun up a narrative that commanded a premium valuation, but once fundamentals started actually deteriorating instead of catching up, that became a big problem.

Investing while hoping a future materializes without being clear how, is riskier.

From DCF to Multiple.

Let's get into the actual math behind tying multiples to the DCF.

The model has different growth periods, assuming a different growth rate for each: a company for the next 5 years is going to grow 15%, then from year 6-10, 12%, years 11-29, and it keeps fading.

You're going to probably say, “Why are you projecting this out to 100 years? No company exists for 100 years!”

But you can pick any period you want, 30 years, 20 years.

One of the biggest criticisms people have of the DCF is the terminal value calculation.

It's funny, because you don't actually need to do one.

The reason people do it is because it's convenient, since people don't want to do DCFs out to 100 years, but it's very trivially easy to do a DCF far into the future.

People say, I can only really understand assumptions for the next 5 years, so let's just do the terminal valuation calculation instead.

That's wrong, because the terminal valuation calculation has embedded in it all of these growth assumptions already: it's basically just another multiple, and you don't know what it's assuming.

If you're uncomfortable assuming 100 years out, don't go 100 years out, go 50, go 30.

But don't put a big terminal value at the end that's like 80% of the valuation and say that's more conservative.

It's not, you're still just assuming the valuation's there, it's just much further out.

The way this model works is you're doing the growth rates on the earnings, projecting them out, discounting them back under different discount rates, and that gets us a valuation.

Take that valuation, divide by current earnings, and you get a P/E ratio, the justified P/E ratio.

This is a DCF, just quoted in multiples instead of a stock price.

The higher the required return, the lower the justified P/E ratio: if I'm paying less for a business, I'm more likely to get a higher return.

I want you to fully understand this idea that a multiple is just making assumptions on a DCF.

The purpose isn't to show you're overpaying for Palantir or SpaceX, it's that there's assumptions embedded in prices that look high, but if you're comfortable with those assumptions, it could be a fine valuation.

The Justified Multiple Table.

What you were just looking at was simplified to one growth rate trajectory.

In the actual model, I modeled six different scenarios, each labeled after its starting growth rate and you can see how they fade.

I like to sensitize everything, because whatever happens in the future for a stock, there's not one valuation.

One pet peeve I have with Wall Street analysts who put out a single target price is these things are better thought of probabilistically: you could look at a worst case scenario of 5%, a best case of 30%, and a lot in between.

This is going to be the most important takeaway from all of this.

This is the justified multiple you can pay for a stock to get various returns at various discount rates, because the discount rate is essentially a return for a stock.

These discount rates assume you bought the entire business and all excess cash flows go directly to you as an investor.

If you do this discount rate at 10%, and it's only growing 5% a year, you could pay up to 17 times multiple today.

If it's growing 20%, fading to 15%, 10%, 7%, 5%, 3%, and you want a 10% return, you could pay a 78x multiple.

If they are able to grow 30% a year for 5 years, which is really high growth and very few companies are able to this but if you’re able to do that, you could pay a 245x multiple.

That's a very high multiple.

And so now maybe you're starting to build a little bit of an intuition, is the multiple of earnings for Palantir actually that expensive?

Look at these numbers if you believe really high growth.

And what is Palantir growing today?

Well, it's growing 80%, and at an accelerated rate.

And so the question is how long you believe this is going to really be able to last.

Because let's say they grow 40% for the next 5 years.

That math would be: (1.4)⁵ = 5.4x revenue growth and they are currently at $6.2bn right now.

So in 5 years, revenues would be ~$33bn.

So roughly speaking, low 40-ish percent growth is already being priced in at today's price.

Then the question is, can they grow more after that above what a market return is?

So if after 5 years, you're still seeing these very high growth rates, then maybe that is an interesting opportunity that can present itself.

But I'm trying to switch the way you think to think in terms of assumptions rather than valuations.

The valuations, multiples, we could calculate these.

The real question is, what discount rate do you want?

Maybe you want 15%.

And then what growth are you comfortable with?

A lot of the stocks you're going to want to be looking at are probably going to be lower growth, and you're going to want at least a 10% discount rate, maybe 15%.

With a scenario of 10% growth, which is still high for a lot of businesses, you're looking at low teens, low single digit to a mid-20s multiple.

The Palantir Case Study.

This is projecting out revenue, with a revenue growth assumption and a net margin assumption, discounted under a required return of 10%.

Here's this justified multiple figure: the DCF being run to get the net present value of all future cash flows today, divided by earnings in this period, $2.3 billion, basically the same as taking the stock price and dividing by earnings per share to get the multiple.

What justifies it is that if you believe these growth rates are actually going to happen, and you discount by 10%, that's the multiple you could pay.

So what is the right multiple to pay?

It's a byproduct of the growth you expect and the discount rate, those two things, that's it.

As you go farther out, the justified multiple drops a lot, because earnings are growing a lot.

Right now it's 185x with these revenue growth assumptions, which I'm not endorsing, but if you believe those, you could pay 185x for Palantir and still get a 10% return.

But that's why I have this other line, the implied multiple of the future at today's market price.

The justified multiple keeps rolling forward the net present value using future current-year earnings.

The implied multiple instead holds steady the valuation, $435bn today, versus the justified multiple's future valuation of almost $900bn, which grows because the business is growing in future value.

Today, that justified price is $441bn.

In 5 years, you could pay $620 billion and get a 10% return, rolling forward the DCF another 5 years.

Each year, the implied multiple drops as earnings grow.

By year 10, it's going to trade at 19x earnings.

But if we fast-forward 10 years, how will investors value it then?

That is what the justified multiple attempts to say: at 10 years out, investors will pay 38x for it, because that investor is running another DCF using the same growth assumptions, just pushed forward another 10 years.

So what's your return look like?

First, this future value divided by the current market cap, $435 billion, gets you 100% appreciation, which amortizes over 10 years to 7.5% annual price appreciation.

The next source is the earnings collected over those 10 years.

Those two together get you a 9.2% return.

Why 9.2 and not 10?

Because the current market cap is $435bn, but the DCF says it should really be $403bn, so we're not actually getting the full 10% buying it today at the higher price, that's what the DCF does when it discounts back.

Warnings.

One, the reason why people don't like growth investing is a lot of companies that say they grow very fast usually disappoint, because the base rate is you can't continue to grow at very high rates for a very long period of time.

So it's a risky assumption, but if you're confident in it, the math supports it.

The other thing is I don't really run math like this when I'm actually valuing a business, but this shows you how to get there.

A lot of times, an exit multiple is a mix of a valuation and a pricing: a valuation is what would I pay for it, a pricing is what would someone else pay for it.

You could say I would pay this justified multiple of 38x for Palantir out into the future, whereas the market at that time is probably not going to pay 38x, maybe 30x, and compare notes there.

I also want you to not be too attached to anything that gets spit out by this model, since it's a super long-term model, so keep gut checking it with the implied multiple.

If the implied multiple 10 years out is 19x, that means you are paying a market multiple assuming this growth takes 10 years to happen, 40% and then 18% for another 5 years.

That's pretty high growth, so take that assumption seriously, because it's going to be hard to hold the stock for 10 years.

Companies that have grown a lot have still had vicious drawdowns.

You could have made pretty good money buying Amazon at $3 a share, but still had a pretty large drawdown, almost a lost decade, maybe 8 years.

It's easy to say you believe in your numbers, but I haven't seen many investors hold a stock that's been flat that long.

So gut check it, don't be too brazen with your growth assumptions.

I think it's much easier to look at valuation in terms of a reverse DCF, since it gives you more of a range of outcomes on the probability curve.

Bringing This Altogether.

A multiple has embedded assumptions in it.

Whether you're aware of them or not doesn't matter, that's already factored into the multiple.

To see what those assumptions are, you run a DCF, because a multiple is just a shorthand for a DCF.

When you are doing a valuation, think in terms of assumptions not valuation.

So what is the right multiple to pay?

It's ultimately the byproduct of the growth you expect and the discount rate you are comfortable assuming in the business.

If you want to see how multiples map to a DCF, I walk you through it step by step in this video below, along with giving you a free Excel for you to download.

For More on What is the Right Multiple to Pay, check out the video below.

YouTube FAQ

Nothing in this newsletter is investment advice nor should be construed as such. Contributors to the newsletter may own securities discussed. Furthermore, accounts contributors advise on may also have positions in companies discussed. Please see our full disclaimers here.