Session 18: Optimal Financing Mix II- The cost of capital approach

Profit Playbook
12 followers
In session 18 of this 36 session corporate finance class, I'd like to offer you a tool
to come up with the optimal mix of debt and equity for your company. It's a cost of capital.
The cost of capital, when minimized, maximizes your value of the firm and in the process
provides us with a way of finding the right mix of debt and equity for any business.
So now we're into the middle of estimating the right financing mix for a company.
In particular, in the last session, we set up the trade-off, the benefits of debt versus the
cost of debt. But we really didn't come up with a way of coming up with the right mix as a specific
number. So we came up with a way of saying this company can borrow more money or less money,
but if I put you on the spot and said, should it be 50% or 80%, you wouldn't have been able
to answer that question. So in this session, I'd like to give you a tool for being able to assess
the optimal mix of debt and equity for the company. And it's a tool you should be familiar
with by this point of this class. Remember the cost of capital? We used it as a hurdle rate
investment analysis, right? Basically, it's that number you need to beat for a project to be a
good project. It's a weighted average of your cost of equity and cost of debt. There's another use
for the cost of capital that I'm going to introduce. The value of an entire business
can be written as the present value of the cash flows to that business discounted back at the
cost of capital. You're saying, so what? Well, let's play a little game. Let's assume you can
keep the cash flows fixed and change the discount rate. In particular, as the discount rate gets
lower, the value gets higher, right? The present value gets higher. And if you can minimize the
cost of capital, you maximize the value of the business. Now, part of you might say, so what?
Remember the objective in corporate finance is to maximize the value of your business.
If you buy into that objective, the right mix of debt and equity for your company can be stated
to be the one that minimizes your cost of capital. So, task then becomes coming up with that mix of
debt and equity that minimizes the cost of capital. So, let me start with an abstract example, a very
simplistic example to illustrate this process. Let's assume that I gave you the cost of capital
schedule for an entire company. What do I mean by schedule? I give you the cost of equity, the cost
of debt, and the cost of capital at every debt ratio from 0 to 90 percent, actually to 100 percent.
So, I'll tell you a story with each number, but you're going to see that this storytelling can
take you only so far. As my debt ratio goes up, my cost of equity will increase. Why is it going up
the way it is? For the moment, I don't have a good answer, but the intuition is as you borrow
more money, the equity earnings will get more volatile. You should demand a higher cost of equity.
As your debt ratio goes up, your cost of debt should probably also go up because people lending
your money will face more default risks. So, the cost of debt does go up. Again, if you ask me why
is it going up the way it is, I wouldn't have a good answer. After all, this is a made-up example.
But if you buy into that cost of equity and cost of debt number, then the cost of capital you see
computed in the second to last column is the number you're going to focus on. Remember, your
objective is to minimize the cost of capital, right? It happens at a 40 percent debt ratio.
Your cost of capital is minimized to 10.14 percent. To show you the payoff in terms of what it does to
firm value, here's what I did. I kept the cash flows fixed because the cash flows are to the
business. They're not cash flows that are affected by how much you borrow and changes the cost of
capital. The point at which your cost of capital is minimized is also the one at which the value
of your business is maximized. Shown as a graph, you can see it peak, the value of the business
peak at a 40 percent debt ratio. That is your optimal debt ratio. This is if computing the
optimal debt ratio is as simple as looking up the schedule, anybody should be able to do it,
right? Because all you need to do is find either the low point in that table or the high point on
this graph in terms of value, and you've got the optimal debt ratio for any company. The challenge
you face with any company though, with a real company, is when you try to do this, you don't
have a schedule, you don't have a graph, you have a point. The only number you actually know is the
and the task you're faced with is how you flesh out the rest of the chart.
That's effectively what I'd like to address for the rest of this session.
So I'm going to try this for Disney, but before I get ambitious, let me nail down the one point I do
know for Disney. Disney's existing debt ratio is about 11.6 percent debt, 88.4 percent equity. In
fact, this is a reproduction of a page you saw earlier in the context of measuring hurdle rates.
At this existing debt ratio, we computed a cost of equity and a cost of debt and an overall cost
of capital of 7.81 percent. So one way to think about the question we're raising right now is,
by changing the mix of debt and equity at Disney, can I make the cost of capital a number lower
than 7.81 percent? Now think of what you need to come up with a cost of capital schedule for Disney.
You need a cost of equity at every debt ratio from zero to 90 percent, right? You need a cost
of debt at every debt ratio from zero to 90 percent. We've actually set up a process earlier
that we can now draw on to answer those questions. And in effect, here's what we're going to do.
Remember how we computed the cost of equity for Disney? We started with an unlevered beta based
on the business it's in. We then levered it using a debt to equity ratio. That's good, right? Because
now if I change my debt to equity ratio, you should be able to tell me what the levered beta
will be at the debt to equity ratio and what the cost of equity will be. That's exactly what I'm
going to do to get the cost of equity at every debt ratio. To get the cost of debt, remember how
I did it for Disney. I used the actual rating, but I also told you how I would have done it if I did
not have an actual rating. I said I'd compute an interest coverage ratio, operating income divided
by interest expense. I'd use that interest coverage ratio to estimate a synthetic rating
and the synthetic rating to get a cost of debt. That's exactly what I'm going to do to get the
I'm going to compute the interest coverage ratio at every debt ratio. Remember, as I borrow more
money, my interest coverage ratio is going to decrease because my interest expenses will climb.
As my coverage ratio decreases, I'm going to estimate a synthetic rating. As I borrow money,
that rating will probably get worse and worse. Use that rating to come up with a pre-tax cost
of debt and an after-tax cost of debt. The rest is pure mechanics. I am going to be able to create
a table that looks just like that hypothetical company I started this discussion with,
and then you should be able to come up with the optimal debt ratio for Disney.
Before we get started, there are a couple of numbers I'd like to nail down. The first is what
the unlevered beta for Disney is as a company. I can use a regression beta and use the average
debt-to-equity ratio over the last five years. That yields a beta of around 1.1119, but remember
I said I don't trust the regression beta very much. I'm going to use the other approach we
came up with for the unlevered beta, where we broke Disney down into five businesses,
looked up the beta by business, and took a weighted average. That unlevered beta of 0.9239
is the beta that I'm going to use for Disney. Incidentally, if you don't buy into the bottom-up
beta approach or you cannot use the bottom-up beta approach, use the regression beta approach
to come up with an unlevered beta. That's going to be the starting point for the cost of equity.
Let's look at the numbers that will drive the cost of debt. The interest coverage ratio, right? In
the numerator, you have operating income. So first, I need to figure out how much operating income
Disney has available. That's exactly what I've done in this table. I've looked at the operating
income in the most recent 12 months, and just for comparison, shown you what the numbers look like
in the previous year. The operating income in the last 12 months, if I convert leases to debt,
is 10,032 million. That's what you're going to see as my base operating income for Disney.
Because I'm going to treat leases as debt, then I have to use the operating income that's adjusted
for leases. You're saying, what do you mean adjusted for leases? If you looked at Disney's
income statement, you're going to see an operating income number. That operating income number
reflects the treatment of lease expenses and operating expense. Because I've capitalized
the leases, here's what I had to do. I had to add the lease expense back and subtract out the
depreciation on the asset I created from the lease. That's the 10,032 million, and that's what
you're going to see in the numerator of my interest coverage ratio. So you ready? Let's get the
process rolling. Let's do the easy half first. Let's first estimate what the cost of equity
will be at every debt ratio from 0 to 90%. At every debt ratio, here's what I do first. I
compute a debt to equity ratio. So let's take an example. Let's say my debt ratio is 20%.
That means I have $20 of debt for every $80 of equity, right? You divide the 20 by the 80, you
get a debt to equity ratio of 25%. I plug the debt to equity ratio into that equation we came up with
for the levered beta. Remember, my unlevered beta stays at 0.9239. My marginal tax rate stays at
36.1%. The only number that's changing is my debt to equity ratio. I get the levered beta at every
debt ratio. I use that levered beta in conjunction with my risk-free rate, which stays at 2.75%
because I'm doing everything in U.S. dollars, and my equity risk premium of 5.76% that I estimated
for Disney to come up with the cost of equity at each debt ratio. Half my table is done. Now let
me turn to the cost of debt because it's a little messier. To show you how I'm going to compute the
cost of debt, I'm going to take a 10% debt ratio and work it through to see the mechanics of what's
going to happen. To get the process rolling, I'm going to start by adding the market value of equity
to the market value of debt. So this is the market capitalization of the company to the 15.96 billion
I estimated in debt, which includes the leases. And the market value that I get for the overall
company is about $13.78 billion. Why do I need that? When I look at a 10% debt ratio, the first
number I need to estimate is how much dollar debt I will need to get to a 10% debt ratio.
And 10% of $137,839 is $13,784 million. So that's my starting point. That's how much debt I will have.
Then if you look at the EBITDA, the DA and the EBIT numbers, notice that they're exactly the
same at a 10% debt ratio as they were at zero. You're saying, how can that be possible? If I'm
borrowing $13,784 million, how come I'm not affecting the operating income? Here's why it's
happening. Whenever you look at assessing capital structure, you do what's called a recapitalization.
What is that? You keep your projects fixed and you change your mix of debt and equity.
The $13,784 million that I borrow goes back into the company and I buy back stock with it. I reduce
equity. In other words, it keeps my operating income fixed because that's the only way I can
isolate the effect of interest expenses or the amount of debt that I have. So in this case,
my EBITDA, DA and EBIT remain the same as they were at a zero percent debt ratio.
Now comes the first difficult number. I need an interest expense and here's how I compute it.
I take the $13,784 in debt and multiply it by my interest rate of 3.15%. There's a problem in what
I'm doing and I'm going to come back and ask you what the problem is. But with that assumption,
I get an interest expense of $434 million. I divide the operating income by the interest expense
and the interest coverage ratio I end up with is 23.10. Now I turn to the next page and I look up
that interest coverage ratio. This is a large market cap lookup table and the rating I have
at that ratio is AAA. The cost of debt I have is 3.15%. I go back to the previous page. I've got
my cost of debt. I can move on. Let me list the steps because there is actually a flaw in this
process. Let's see if you can find it. I start off by estimating the dollar debt, right? The 10%
debt ratio times the value of the firm. I multiply the dollar debt by the interest rate to come up
with the interest expense. I compute an interest coverage ratio using that interest expense. I use
the coverage ratio to come up with the rating. Use the rating to come up with an interest rate.
Notice a problem here. I'd need an interest rate in step two and I need the interest rate to get
the interest expense to get back to the interest rate. It's a classic chicken and the egg problem.
Unlike the chicken and the egg problem, to which I don't think there's a known solution,
let me offer you a solution. To see what the solution is, I'm going to move to a 30%
debt ratio and work through the numbers. You can see exactly what the problem is and how I try to
fix it. Let's assume that I have the cost of debt for Disney at a 20% debt ratio. At a 20% debt
ratio, let's say that the rating for Disney stays AAA and the cost of debt is 3.15%. Now the question
I'm trying to address is what will the cost of debt be at 30%? The first step in this process
I have to estimate how much dollar debt I will have at 30%. That dollar debt is going to be about
$41.3 billion. Effectively, I've added another $13.784 billion because that's what 10% debt does.
The EBITDA, the depreciation, the EBIT stays fixed because remember what we said.
We're assuming that this is a recapitalization. The project stays the same. Then I get to the
interest expense. My first go around, remember I don't have an interest rate yet, so I used the
3.15% I knew from the previous level of debt to compute my interest expense and that gives me an
interest expense of $1.302 billion. I use that interest expense to come up with an interest
coverage ratio of 7.70. Now turn back to the previous page. An interest coverage ratio of 7.70
gives me a rating of AA and a cost of debt of 3.45%, right? We have a problem. Here's the problem.
The interest expense was computed using a 3.15% interest rate and now I'm claiming that the
interest rate should really be 3.45%. But here's the fix to the problem. Recompute the interest
expense. The interest expense with a 3.45% interest rate applied to the total debt of
$41.3 billion is $1.427 billion. With that interest expense, my coverage ratio decreases to 7.03.
I turn back to the previous page very gingerly because I'm afraid I might have slipped another
notch. That didn't happen. Here the rating stays AA and because it stays AA, the cost of debt is
3.45%. I'm done. That's exactly what I have to do at every debt ratio and I have to do as many
iterations as I need to to get to this point of steady state. With the interest rate I used to
compute an interest expense is exactly the same interest rate I claim the company will have at
debt ratio. So I have a cost of debt of 3.45% pre-tax at a 30% debt ratio. I multiply by 1
minus the marginal tax rate of my after-tax cost of debt. I do this for every debt ratio and you
see the numbers captured in this table. If you look at this table, we have the ratings at every
cost of debt based on the interest coverage ratio and as the ratings decline, the cost of debt goes
up. And I multiply that pre-tax cost of debt by 1 minus the tax rate. Now there's a little quirk
here that I want to draw your attention to. Until you get to about a 60% debt ratio, notice that my
marginal tax rate is 36.1%, the number we've been using so far. At a 70% tax rate, something seems
to happen. I have a lower tax rate. That's not a good thing because it actually makes my after-tax
cost of debt higher. So to show you what's happening between 60% and 70%, let me focus
in on those two debt ratios. At a 60% debt ratio, my interest expense is 9.5%. That's a lot of
interest expense, but it's still less than my operating income. I'm able to claim my entire tax
benefit. At a 70% debt ratio, my interest expense actually exceeds $11 billion. That's higher than
my operating income. And here's what I do. I cap your tax savings at the $10,032 million in operating
income. What do I mean by that? I allow you to get 36.1% of the $10,032 million as your tax savings
and that tax savings is $3,622 million. But now you're done. Any interest expense above that,
you no longer get the tax savings. If I divide the $3,622 by the $11,096 million, you get the
tax rate of 32.64% that you see at the 70% debt ratio. Things get worse at 80% and even worse at
90%. So effectively, the after-tax cost of debt will climb much more steeply after 70% because
my tax benefits have gone away. So I have a cost of equity from my levered beta. A cost of debt
based on the rating and the tax rate, I bring them together in a table that looks very much
like that hypothetical company's table we started this discussion. Remember again,
what we're looking for is the lowest cost of capital. For Disney, that number, just as it was
for the hypothetical company, is 40%. Don't take that as some rule of thumb you can use. The
optimal debt ratio for all companies is not always 40%. But at a 40% debt ratio, Disney's cost
of capital is 7.16%. That's 0.65% lower than its existing cost of capital. That is the trade-off
we're looking at, and we look at it as a graph. It doesn't quite resemble the U-shaped graph
because initially, the decline in the cost of capital is relatively, you know, it's mild. But
once you get to 40%, notice there's an inflection point, your cost of capital starts climbing.
This is closer to reality, and one of the things I take out of this graph is if you're going to
make a mistake, it's better to be under-levered, have too little debt than too much, and you can
see why. Because once you get past that tipping point, look at how quickly your cost of capital
climbs. That's something we've got to keep in mind when we make recommendations about what Disney
should do about its excess debt capacity. In fact, just as a final graph, I'd like to show you
another cost of capital graph from way back in time. The very first time I assessed Disney for
one of my books was in 1997. It's the cost of capital graph I came up with for Disney then.
Notice the one little difference between this graph and the one I showed you just now.
This graph, if you look at it, 60%, 70% debt ratio has a kink in it. It kind of goes up and
then goes back down. There's a reason for that. The reason is that our cost of debt moves
discontinuously. What I mean by that is for your cost of debt to change in the approach that I've
just described, the interest coverage ratio has to put you at a different rating. If that doesn't
happen, it's entirely possible that you could get one of these quirks where the cost of capital has
a kink in it, where initially it goes up and then goes back down. I wouldn't lose sleep about it.
The optimal debt ratio for Disney is not going to change just because you make the kink go away,
but it's a discontinuity in the process by which you estimate cost of debt that causes a kink cost
of capital. It's something you can fix if it really bothers you, but the process for estimating cost
of capital is a simple one if you go back to basics. In the next session, we will build on
this optimal cost of capital and talk about why companies might want to go to the cost of capital,
but for this session, that's about it. Thank you very much for listening.
