Session 20: Optimal Financing Mix IV - Wrapping up the Cost of Capital Approach

In this session 20 of a 36 session corporate finance class, I'd like to complete the discussion of the cost to capital approach to coming up with the optimal mix of debt and equity for your company. After talking about how to extend this approach to commodity companies, private businesses, and emerging market companies, I'd like to also talk about the determinants of that optimal mix and why it might vary across different businesses. In the last two sessions, we used the cost to capital approach to come up with the optimal debt ratio for a company, in this special case, Disney. And then followed through by looking at why moving to that optimal made sense for Disney stockholders. And also looked at what if something went wrong and whether Disney could use that excess debt capacity to take investments. In this session, I'd like to wrap up the cost to capital approach by extending it to my remaining companies, including a private business. So here's where we are in this process. We started by looking at the trade off on debt and equity. We had two sessions we spent on the cost to capital, and this is going to be the last session using the cost to capital approach as a way of coming up with the optimal debt ratio or a financing mix for a company. Now, here's the first extension I want to talk about. There are two significant limitations in the cost to capital approach that I just described. The first is that I don't factor in what I call indirect bankruptcy costs. Remember, indirect bankruptcy costs kick in because once you drop below a certain level of default risk, if you become too risky, customers stop buying your products, your revenues might suffer, your operating income will suffer. If you remember what we did for Disney at its different debt ratios, we kept the operating income fixed. You could argue that that is unrealistic. That if Disney, in fact, goes out and borrows more money, and its rating drops below triple B, even to single A or from any level, you could argue that maybe there will be operating income effects, which I haven't factored in. So that's the first limitation of a traditional cost to capital approach, is I'm not factoring in the effect of declining ratings on operating income. The second is that the numbers I'm using are static numbers. I'm using 2013 numbers. And to the extent that they can change, my optimal debt ratio will also change. So what I'd like to do first is go back to the Disney analysis and think about ways in which I can bring an indirect bankruptcy cost into the process. And here's the simplest way I've thought of doing this. Let's assume that the way indirect bankruptcy costs kick in is as your rating drops, your operating income is affected. In the case of Disney, let's assume that until you get to single A, in terms of bond ratings, there is no indirect bankruptcy cost. As a manufacturing company, people don't care whether you're double A, triple A, or single A. But that once you drop to A minus, you start to see the first impact. And it's very small initially. And as your rating keeps dropping, your operating income starts to reflect it. I've actually listed out three possible scenarios from highest to lowest possible impacts on operating income, a high, a medium, and a low. And for Disney, I actually went with the middle column because I think that the indirect bankruptcy costs are not going to be at the high end or at the really low end of the spectrum. There are going to be some bankruptcy costs. So the middle column is the column that I'm going to use in recomputing the optimal debt ratio for Disney. You're saying, what does that even mean? Remember the original cost of capital approach? All I cared about was the cost of capital. I looked to see where that cost of capital is minimized. In the enhanced cost of capital approach, here's what I'm going to do in addition to allowing the cost of capital to vary. I'm also going to allow the operating income to change as my rating changes. So two things are changing in my analysis. One is the cost of capital. The other is the operating income. I can no longer focus on just where the cost of capital is minimized because the point at which my cost of capital is minimized might not be the point at which the value of the business is maximized because the operating income is also changing. So the very last column is the column I'm going to focus on. And rather than looking to see where the cost of capital is minimized, I want to see where the value of Disney as a company is maximized. I have some good news here. My optimal debt ratio remains at 40% even if I include enhanced bankruptcy costs. And the reason is simple. Remember, the rating for Disney is still single A at a 40% debt ratio. But here's what indirect bankruptcy costs do to my analysis. Remember that inflection point where my cost of capital started going up after 40%? I said watch out for it. Now you have to really watch out for it because it gets even more dangerous. Because once you cross that inflection point, not only does your cost of capital start to go up, but your operating income starts to collapse. So it's one more reason why you want to stay as far away as you can from that point, which might be 42%, 43%, or 43% debt. So maybe you might want to stop at 35% debt because of the indirect bankruptcy costs built into the process. For most companies, when you bring in indirect bankruptcy costs, you'll find that the optimal actually shifts downwards. Which makes sense, right? If you have more indirect bankruptcy costs to worry about, you should borrow less money. So that's the first enhancement I wanted to talk about. Now let me get back to the traditional cost of capital approach, which I'm going to apply to the remaining companies in my mix. Let me start off by computing the optimal debt ratio first for Tata Motors. To compute the optimal debt ratio for Tata Motors, I did exactly what I did for Disney. I computed 11 beta at each debt ratio, a rating and a cost of debt at each debt ratio, and a cost of capital at each debt ratio. I computed the optimal debt ratio by looking at where the cost of capital is minimized, and it looks like the optimal debt ratio for Tata Motors is about 20% debt, 80% equity. At their actual debt ratio of 29%, Tata Motors looks over level. Now here's where being part of a family group kind of mixes the equation, makes it more difficult to make an assessment. Maybe Tata Motors does have too much debt, but if you're a banker lending to Tata Motors, my guess is that you're implicitly assuming that the family group will step in and bail you out if you get into trouble. Dangerous assumption because it's an implied guarantee, not an explicit guarantee. It doesn't mean that bankers won't lend money. So I can see why Tata Motors is borrowing more than it should, because it's falling back on the family group. In fact, I did this analysis for four other Tata companies, or three other Tata companies, Tata Steel, Tata Chemicals, and Tata Consulting Services, and here's what I found. I found that Tata Motors, Tata Steel, and Tata Chemicals were all over level, and Tata Consulting Services was under level. I know this might sound conspiratorial, but here's what looks like it's, from the outside at least, here's what I see happening. I see Tata Motors, Tata Steel, and Tata Chemicals borrowing money using Tata Consulting Services' debt capacity implicitly and using that debt capacity. It's a dangerous game that's being played, but it's a game that gets played in family group companies all the time. So that was my first assessment, Tata Motors looks over level. Then I moved on to Vale. I computed the optimal debt ratio for Vale to be about 30%, and it looks over level because it has a debt ratio that's actually higher than 30%. But that's using 2013 operating income, just as I did for Tata Motors and Disney. But here's what worries me about using 2013 numbers for Vale. Vale is an iron ore company. It's a commodity company, and 2013 was not a good year for commodity prices. Its earnings reflected that. If I normalize earnings, which sounds fancy, but if I take an average earnings across the last three or four years, and I use that normalized income instead of last year's income, the optimal debt ratio for Vale jumps to 50%. I'd be more inclined to go with that normalized number here than the actual earnings, because as a commodity company, I need to borrow money based on what I make in a regular year, in a normal year, not what I make in a bad year, or not what I make in a good year. In fact, as a general rule, when you compute the optimal debt ratios for commodity companies, try to do what I did for Vale. Look at the average operating income across time. Use that average operating income in assessing the optimal debt ratio. You don't have to average everything else. It's the numbers that are volatile across time that you have to average out, and the operating income is going to be the primary culprit. Now, let's think about applying this approach to a growth company, and I'm going to try this on Baidu. Baidu is a successful company. It's got a high market cap, and that might, in fact, be the reason why, when you look at its optimal debt ratio, the optimal debt ratio you see for Baidu is very low. It's only 10%. In fact, it's probably a number between 0% and 10%. Why is the optimal so low? Here's the answer. Baidu is a successful company, but it's a growth company. Its market cap reflects it. Its operating income as a percentage of its market cap is a low number. So even as it moves to 10% and beyond, that might sound like a low debt ratio, but it finds its rating drop off very quickly because it doesn't have the operating income to service its debt. In fact, if you look at Baidu's actual debt ratio, 5.23%, that's actually pretty close to where it should be, given the fact that it doesn't have much debt capacity. So as a general rule, when you look at young growth companies, don't be surprised to see optimal debt ratios of 10% or even 0%. Given the lifecycle judgment we made earlier, companies that are that young should not be borrowing money, and the cost-to-capital approach backs up that conclusion. So the cost-to-capital approach actually applies even to young growth companies, which brings me to my final optimal debt ratio computation, at least in the context of regular companies, and that's BookScape. It's a private company, right? The starting point for my optimal capital structure analysis is to estimate what the market value of the firm is, and with a private company, I don't have that number. So here's how I got it. I took the estimated present value of lease commitments. That I was able to get at 12.136 billion. But to get my market value of equity, I took the net income, which I was able to get even though it was a private company, and I applied a multiple to that number. Multiple based on what? What publicly traded book companies traded, which was about 20 times earnings. That allows me to estimate a market value of equity. That estimated value of the companies, what I'm going to compute my debt ratios, my dollar debt, and my cost-to-capital on. And based on that assessment, the optimal debt ratio for BookScape works out to be 30%. At its actual debt ratio of 27.81%, BookScape looks fairly level. It looks like it has the right amount of debt. I do have to add, though, that if it had come out as under-levelled, unlike public companies, where I would put pressure on managers to move towards their optimal, with a private business owner, if the owner said, look, I don't want to borrow money, it makes me uncomfortable, my response is, it's your company. And if that's what you want to do, that's fine with me. So those are five very different companies, and we've come up with optimal debt ratios using a traditional cost-to-capital approach for each of them. Now, let me talk a little bit about coming up with the optimal debt ratio for a bank. To see how it's going to work, let's think of an old-fashioned bank. What's an old-fashioned bank? Here's what it does. It takes deposits from individuals and it lends them out. It makes money on the spread. Let's assume this company, this bank, has $100 million in loans outstanding that it has made, and that right now it has a book value of equity of $6 million, and let's assume that that is its regulatory capital as well. So it has a regulatory capital ratio of 6%, and its loans outstanding are $100 million. Let's assume it's a high-growth bank. It expects those loans to increase to $150 million next year, and it wants to increase its regulatory capital ratio to 7%. Why? Because it wants to be safer. It doesn't want to get into a crisis. Think about how much it will need as regulatory equity next year. $150 million will be the loans outstanding. 7% of that is $10.5 million. It right now has $6 million. That effectively means it's got to come up with an extra $4.5 million in equity next year. What does that mean? If this bank came to me for some advice, my suggestion first would be to look at their net income. If they have $6 million or $8 million in net income, they're home free because all they need to do is make sure their retained earnings is $4.5 million. So their dividend policy would have to be adjusted to give them enough additional book equity or retained earnings of $4.5 million. If they do not have enough net income to do that, then they have to raise fresh equity. So for a bank, when they ask, am I at my right financing mix? The question they're really asking is, should I be issuing new equity? And to answer that question, my suggestion is go back to the regulatory capital and build up to that. That's effectively one way to think about the optimal debt ratio for a bank, or the optimal financing choices for a bank is to think in terms of regulatory equity. Now let's get back to the optimal debt ratio. The five very different companies we've looked at. If you look at why the debt ratios vary, and they do vary, right? They go from 0% for Baidu to 40% for Disney. Here's the first reason. It's your marginal tax rate. Remember the marginal tax rates vary across these companies. And as the marginal tax rate goes up, your optimal debt ratio should go up. In fact, I took each of my companies and looked at what would happen to the optimal if I let the marginal tax rate go from zero to 10 to 20 to 30 to 40 or higher. Notice that as my tax rate rises, my optimal debt ratio rises for every company. Conversely, if I make my marginal tax rate zero, the optimal debt ratio for every one of my companies, no matter where they are in the world, goes to 0%. The biggest single benefit of debt is the tax benefit. And if you take that away, companies will borrow a lot less money. Here's the second factor. I looked at how much these companies had as EBITDA and operating income as a percentage of market value. The higher this ratio, the more a company should be able to afford to borrow. So if you take Disney, for instance, its EBITDA as a percentage of enterprise value is about 9.35%. Its optimal debt ratio is 40%. If Disney were able to double that percentage, their optimal debt ratio would be much higher. If you take a look at companies with optimal debt ratio of 70, 80, or 90%, it's usually because they generate huge cash flows as a percentage of market value. Conversely, if you look at Baidu as a company, its EBITDA as a percentage of firm value, of enterprise value, is less than 4%. Its optimal debt ratio reflects that judgment. It's only 10%. So the more operating income you have as a company relative to your market value, the higher the optimal debt ratio can be for your company. So your tax rate matters. How much you generate in cash flows matter. Your risk matters as well. The more operating risk there is in your business, and it shows up through your unlevered beta in your rating, the lower your optimal debt ratio will be. And that makes sense, right? Riskier companies for any given level of operating income should borrow less money. So these three variables, the tax rate, your cash flows, and your operating risk, all reflect company choices or company fundamentals. There's only one macroeconomic variable that seems to affect your optimal debt ratio. And that is what the market price for risk is in the equity market relative to the market price of risk in the debt market. Let me back up. Remember we talked about equity risk premiums in one of our earlier sessions? That's the price of risk in the equity market. And the blue line on this graph, the line towards the top, is actually the equity risk premium over time. The black line is actually the default spread for a BAA-rated bond. And in a rough sense, it's a measure of the price of risk in the bond market. Most of the time, the two move together, right? But there are times in history where they haven't. Late 1990s, for instance, the equity risk premium drops while the bond default spread rises. If you're a company in 1999 trying to raise money, you could have gone to the equity market where people were charging almost nothing for risk. The risk premium is 2%. Or gone to the bond market and paid a high default spread. It's a no contest, right? Given a choice, you're gonna raise equity. Late 1990s were a golden age for equity financing. Then you get to the first part of the last decade, post 9-11, where Alan Greenspan and the Federal Reserve essentially made the decision to keep the default spreads in the bond market low. And while they were doing that, the equity risk premium stayed high. This is the golden age of the leveraged buy. You could borrow money at relatively low rates and buy equity at relatively low prices. So when equity risk premiums stay high and default spreads drop, you're gonna see companies use more debt. When equity risk premiums drop and default spreads rise, you're gonna see companies use more equity. That's the only macroeconomic variable that drives that mix of debt and equity for a company. So when you get a chance, take your company through this process. Come up with the optimal debt ratio using the cost of capital approach. Then look at what'll happen to your value as a business if you move to the optimum. Ask the what if question. What if my operating income drops? See how much buffer you've built in. And then ask yourself whether in fact your company should think about moving to the optimum. That is a good starting point for assessing the financing mix for a company. Thank you very much for listening.