[MUSIC] We just learned how to estimate cost of capital and how to use cost of capital to figure out whether a company is generating economic value or not, right. That is very useful, and it seems, actually, quite easy, but here comes a problem. Most real world companies are going to have different division, right? You're not investing only in one type of product, you might be investing in many different products. That's true even for PepsiCo, right. Most of PepsiCo's investments are related to soda, right, soft drinks. But PepsiCo also has other food business. The general point is that if a company has different divisions, the company wide cost of capital is generally not the right discount rate for all of them. Okay? Remember the initial idea. Discount rates have to reflect the specific risk of each project. Right? In this case, the discount rate has to reflect the specific risk of each division. Right? And if the company has many divisions then all hell breaks loose. The answer maybe no. Maybe that the company-wide cost of capital is not doing the right job. So now what I want to talk about is how would we do it, right? If we have this problem, if the division is different from the company, how can we do this calculation? Okay. The example I'm going to use is Altria, which is a conglomerate that produces mostly tobacco. Okay. Most of their business is to sell cigarettes, smokeless tobacco and cigars, mostly cigarettes. That's the big business. Both in the US and abroad, okay. But they also have some other smaller divisions. In particular they have a wine division that we're going to talk about here. And also some financial services. This is the wine division of Altria. It makes wines in Washington state, the division is called Ste.Michelle Wine Estates which you might have seen before, especially if you live in the US, these are broadly, these wines are sold very generally in supermarkets and all that, so you might have seen them. So these wines are made in Washington state, okay? And now they are made by this company called Ste.Michelle Wine Estates. They also owned by some other wineries around the world but this is the big business, okay? What we want to do then, suppose if we are Altria, right? The CFO of Altria has to figure out whether the wine division Ste.Michelle Wine Estates is generating economic profit or not. Okay? So how do we do that? How do we compute EVA for the wine division? Right? The first thing we need is data on profits and assets, separated by division. Right? And this data might be difficult to come by. Companies are not necessarily required to report this data. So some of them do, others don't, okay. If you are inside the company you are definitely going to have that information. Even at a more granular level, even for specific projects and all that. In the case of Altria, they were nice, and they actually report data separately for the wine, for different divisions. In particular we are using here the wine and the tobacco division. We have operating profit and assets separately by division. What we don't have is cash separated by a division here, okay? It turns out that Altria does not hold that much cash, so we're just going to assume for this calculation that cash is equals to zero, okay? Remember that in general, you want to deduct cash from all from your assets to figure out operating assets, okay. We do have OPAT, so all we have to do is to tax the operating profit. These taxes are 30% then your OPAT would be $57.8 million, okay? So in 2014, the wine division of Altria generated $57.8 million in profit, right? And the question that the CFOis trying to figure out is this or not? Right? Is this a sufficient profit for the wine division or should be looking for more? Does the division have to perform better, right? Think about the following that's really going to get to the fundamental question we are talking about in this session, okay? We can get beta for Altria, that's very easy. Just go to Capital IQ or do a regression, okay. You'll find out that Altria also does not have a very high beta. Similar to PepsiCo, Altria's beta is 0.5, okay. We can use the range and all that, the bottom line is that Altria will have a lower beta. Maybe between 0.4 and 0.7, like PepsiCo. Is this the right beta to use? If you're trying to answer the question we just posed, which is to figure out the performance of the wine division, is that the right beta? Which risk is this beta reflecting? The answer to put in very simple words is that 0.05 is a tobacco beta. How do we know? Because most of the company is tobacco, right. If you look at the market value of Altria, it's going to be driven mostly by the tobacco division. If you look at the table that we just worked with, most of the assets are invested in tobacco. Most of the profits come from tobacco. Most of the equity value is going to reflect tobacco. There is no guarantee that if we use 0.5, we are actually going to be measuring a wine beta, okay. So now what we need to think about is, then how do we do it, right? It's really very important because EVA, it is useful to figure out EVA for the company as a whole but EVA is even more useful when you think about individual projects, when you think about divisions, right? You can use EVA for executive compensation for example, right. If the wine division is doing great, then the chief manager of the wine division might get paid more, might get more stock for example. But to determine that, you have to figure out the cost of capital for the division. And we cannot directly measure it. Why? Because the wine division does not trade separately in the stock market. You cannot find the stock price for the wine division. There's no way you can estimate the beta even if you had the most sophisticated statistical tools in the world, you're not gonna be able to estimate beta because you don't have the data. So there two approaches that we use here in corporate finance. One is to take a range of similar companies, and use an industry beta, right? So that's the beta that is going to be averaged out across many firms in a related industry. The second one is what we call a pure play approach, which is to try to find a singular company that is publicly traded. So let me give you example here for Ste.Michelle for Altria's wine division. We can start from a beverage beta, right? A beverage beta is definitely going to be better than a tobacco beta, right? Using this table here, you'll see that the beverage beta, using 35 company's in this case, is 0.95, right? So it's a beta that is a lot closer to one, than the tobacco beta, okay? We can also try a pure-play approach. Which is to, in this case, we're looking for the wine beta. So what we need to search for is a company that mostly makes wine. A company whose main business is wine, and that is publicly traded in the stock market. In this case we are lucky. We can use a company called Constellation Brands, okay? Which produces wine all over the world, including the US, Australia and many other parts. And we can get data for Constellation Brands. We can go to Capital IQ or Yahoo Finance for example and you will find that the beta is around one. The beta is 1.02 in this case. So think about this we had the option of using the tobacco beta, right. We had the option to use the tobacco beta. But it actually turns out the beta, a wine beta, wine WACC, would actually turn out to be higher, right? Because, remember, the higher the beta the higher the WACC, right? So if we use a beta of 1 here, we are going to end up with a required return on equity of 8% right? Instead of six and a half percent which we would have obtained if we were using the tobacco beta. Okay? So just a couple more data points here to compute the WACC to go from required return on equity to the WACC. We have to add the debt to value and the cost of debt, right? We take those from Altria. Those are company wide values. Again there could be a difference in the cost of debt from the wine division to the tobacco division but it's less likely to be the case. Going back to the idea we discussed before, a debt securities are safer. The equity gets the residual cash flows. So the cost of that, the required return on that for the tobacco division should be closer to the required return on that for the wine division and here we don't have a good data to get a better estimate so let's just use that, okay? As it turns out, Altria does not have a lot of leverage anyway so that's not even if we are getting this numbers slightly wrong, It's not gonna throw off our estimate that much. Okay? So if we apply the WACC formula, we're going to find a cost of capital of 7.4%. Okay? So the cost of equity, the required return of equity is eight. The cost of capital is 7.4%. Okay? And now, we can go back and estimate EVA. Right? Using, for example, 7.4 we would deduct. We have the OPAT, we have the operating assets. We deduct the WACC times operating assets from the OPAT. We get an EVA of minus 6.3 medium. Okay? What does this mean? Right? It means that looking at that year in isolation, 2014, the wine division did not generate sufficient economic profits, okay? The required profits would be 7.4% of the asset which are bigger than the OPAT minus taxes which is our estimate for the profits that the wine division actually generated. And go back to what we've been discussing. Which mistake would you have made if we had used the tobacco beta? Suppose we had used the 0.5 instead of one for the beta? You would have ended up with a higher EVA. Right? Essentially, you would be underestimating risk, because beta measures risk. So 0.5 is very different from one. Right? Because you're using a low beta, you would have ended up with a lower WACC, and you would be overestimating EVA. For the company it matters a lot, if you're using EVA for performance evaluation, for compensation, you have to try to get it right. So even if this calculation isn't perfect, it's definitely better than if we had used 0.5, okay? How would we use EVA? So suppose that we have checked this calculation and really made sure they're not of that all the numbers are solid and that we find out that the wine division is, in fact, generating negative EVA. What does that mean? Should we go ahead and restructure this division? Should we sell it off? What should Altria do? Here you have to be a bit careful. Notice that when we measure EVA, we are measuring current profits. EVA by the definition, tells you if a company generated economic profits in a given year or if a division generated economic profit in a given year. So what might be happening is that perhaps the wine division is expected to grow a lot, and to generate higher profits in the future. So, having negative EVA today doesn't mean you have to sell off the wine division. What it does mean is that you have to make sure the wine division, the performance of the wine division, is going to pick up. If that never happens, if EVA never turns positive, then at some point the restructuring is necessary. Essentially this means that, that division is not creating economic value. Something has to be done. So it's very useful information for a CFO, for a CEO, for the company as a whole to use EVA and be able to figure out whether the division or a project is contributing in economic value or not.